ggcristian commited on
Commit
d02c1e3
·
1 Parent(s): 73cf928

November release: we add new (big) models and mark the release of TuRTLe v2 codebase

Browse files
config/constants.py CHANGED
@@ -64,7 +64,7 @@ SCATTER_PLOT_X_TICKS = {
64
  "ticktext": ["8", "14", "32", "72", "200", "700"],
65
  }
66
 
67
- DEFAULT_MAX_PARAMS = 700
68
  DEFAULT_TASK = "Spec-to-RTL"
69
  DEFAULT_BENCHMARK = "All"
70
  DEFAULT_MODEL_TYPE = "All"
 
64
  "ticktext": ["8", "14", "32", "72", "200", "700"],
65
  }
66
 
67
+ DEFAULT_MAX_PARAMS = 1000
68
  DEFAULT_TASK = "Spec-to-RTL"
69
  DEFAULT_BENCHMARK = "All"
70
  DEFAULT_MODEL_TYPE = "All"
config/model_metadata.py CHANGED
@@ -7,7 +7,7 @@ class ModelMetadata:
7
  url: str # HF model card
8
  params: Optional[float] # in B
9
  model_type: Literal["General", "Coding", "RTL-Specific"]
10
- release: Literal["V1", "V2", "V3"] # release of the leaderboard for which the model was included
11
  model_arch: Literal["Dense", "Reasoning"] # to distinguish between reasoners and non-reasoners
12
 
13
 
@@ -110,18 +110,24 @@ MODELS = {
110
  "https://huggingface.co/openai/gpt-oss-120b", 120, "General", "V3", "Reasoning"
111
  ),
112
  "Seed-OSS-36B": ModelMetadata(
113
- "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct", 36.2, "General", "V3", "Reasoning"
114
  ),
115
  "Qwen3-8B": ModelMetadata(
116
- "https://huggingface.co/Qwen/Qwen3-8B", 8.2, "General", "V3", "Reasoning"
117
  ),
118
  "Hermes-4-14B": ModelMetadata(
119
- "https://huggingface.co/NousResearch/Hermes-4-14B", 14, "General", "V3", "Dense"
120
  ),
121
  "Hermes-4-14B-Reasoning": ModelMetadata(
122
- "https://huggingface.co/NousResearch/Hermes-4-14B", 14, "General", "V3", "Reasoning"
123
  ),
124
  "Gemini 2.5 Flash (Medium)": ModelMetadata(
125
- "https://huggingface.co/google", None, "General", "V3", "Reasoning"
 
 
 
 
 
 
126
  ),
127
  }
 
7
  url: str # HF model card
8
  params: Optional[float] # in B
9
  model_type: Literal["General", "Coding", "RTL-Specific"]
10
+ release: Literal["V1", "V2", "V3", "V4"] # release of the leaderboard for which the model was included
11
  model_arch: Literal["Dense", "Reasoning"] # to distinguish between reasoners and non-reasoners
12
 
13
 
 
110
  "https://huggingface.co/openai/gpt-oss-120b", 120, "General", "V3", "Reasoning"
111
  ),
112
  "Seed-OSS-36B": ModelMetadata(
113
+ "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct", 36.2, "General", "V4", "Reasoning"
114
  ),
115
  "Qwen3-8B": ModelMetadata(
116
+ "https://huggingface.co/Qwen/Qwen3-8B", 8.2, "General", "V4", "Reasoning"
117
  ),
118
  "Hermes-4-14B": ModelMetadata(
119
+ "https://huggingface.co/NousResearch/Hermes-4-14B", 14, "General", "V4", "Dense"
120
  ),
121
  "Hermes-4-14B-Reasoning": ModelMetadata(
122
+ "https://huggingface.co/NousResearch/Hermes-4-14B", 14, "General", "V4", "Reasoning"
123
  ),
124
  "Gemini 2.5 Flash (Medium)": ModelMetadata(
125
+ "https://huggingface.co/google", None, "General", "V4", "Reasoning"
126
+ ),
127
+ "Kimi K2 Instruct 0905": ModelMetadata(
128
+ "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905", 1000, "General", "V4", "Dense"
129
+ ),
130
+ "DeepSeek V3.1 Terminus": ModelMetadata(
131
+ "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus", 685, "General", "V4", "Reasoning"
132
  ),
133
  }
data_processing.py CHANGED
@@ -69,7 +69,7 @@ def filter_leaderboard(task, benchmark, model_type, search_query, max_params, st
69
  subset = subset[subset["Model"].str.contains(search_query, case=False, na=False)]
70
  max_params = float(max_params)
71
 
72
- if max_params < 695: # when re-setting the max param slider we never reach 700 again xd
73
  subset = subset[subset["Params"] <= max_params]
74
  else:
75
  subset["Params"] = subset["Params"].fillna("Unknown")
 
69
  subset = subset[subset["Model"].str.contains(search_query, case=False, na=False)]
70
  max_params = float(max_params)
71
 
72
+ if max_params < 995: # when re-setting the max param slider we never reach 1000 again xd
73
  subset = subset[subset["Params"] <= max_params]
74
  else:
75
  subset["Params"] = subset["Params"].fillna("Unknown")
results/aggregated_scores_icarus.csv CHANGED
@@ -1,4 +1,6 @@
1
  Model,Agg S2R,Agg MC,Agg VerilogEval S2R,Agg VerilogEval MC,Agg RTLLM,Agg VeriGen
 
 
2
  Gemini 2.5 Flash (Medium),63.55,69.84,68.27,69.16,47.99,76.02
3
  DeepSeek R1-0528,76.79,78.86,78.84,79.65,70.04,71.64
4
  DeepSeek R1,75.53,76.99,77.67,77.55,68.49,71.92
 
1
  Model,Agg S2R,Agg MC,Agg VerilogEval S2R,Agg VerilogEval MC,Agg RTLLM,Agg VeriGen
2
+ DeepSeek V3.1 Terminus,71.79,76.57,74.92,76.83,61.48,74.2
3
+ Kimi K2 Instruct 0905,68.72,71.77,70.82,72.44,61.79,65.63
4
  Gemini 2.5 Flash (Medium),63.55,69.84,68.27,69.16,47.99,76.02
5
  DeepSeek R1-0528,76.79,78.86,78.84,79.65,70.04,71.64
6
  DeepSeek R1,75.53,76.99,77.67,77.55,68.49,71.92
results/aggregated_scores_verilator.csv CHANGED
@@ -1,4 +1,6 @@
1
  Model,Agg S2R,Agg MC,Agg VerilogEval S2R,Agg VerilogEval MC,Agg RTLLM,Agg VeriGen
 
 
2
  Gemini 2.5 Flash (Medium),63.27,70.19,67.75,69.55,48.49,76.02
3
  DeepSeek R1-0528,75.83,78.07,77.64,78.78,69.85,71.64
4
  DeepSeek R1,75.78,75.99,78.04,76.42,68.31,72.05
 
1
  Model,Agg S2R,Agg MC,Agg VerilogEval S2R,Agg VerilogEval MC,Agg RTLLM,Agg VeriGen
2
+ DeepSeek V3.1 Terminus,71.35,75.31,74.22,75.43,61.89,74.2
3
+ Kimi K2 Instruct 0905,67.8,71.78,69.72,72.46,61.46,65.63
4
  Gemini 2.5 Flash (Medium),63.27,70.19,67.75,69.55,48.49,76.02
5
  DeepSeek R1-0528,75.83,78.07,77.64,78.78,69.85,71.64
6
  DeepSeek R1,75.78,75.99,78.04,76.42,68.31,72.05
results/results_icarus.json CHANGED
@@ -1,4 +1,554 @@
1
  [
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  {
3
  "Model": "Gemini 2.5 Flash (Medium)",
4
  "Model Type": "General",
@@ -7,7 +557,7 @@
7
  "Result": 92.05,
8
  "Model URL": "https://huggingface.co/google",
9
  "Params": null,
10
- "Release": "V3",
11
  "Thinking": "Reasoning"
12
  },
13
  {
@@ -18,7 +568,7 @@
18
  "Result": 79.57,
19
  "Model URL": "https://huggingface.co/google",
20
  "Params": null,
21
- "Release": "V3",
22
  "Thinking": "Reasoning"
23
  },
24
  {
@@ -29,7 +579,7 @@
29
  "Result": 70.38,
30
  "Model URL": "https://huggingface.co/google",
31
  "Params": null,
32
- "Release": "V3",
33
  "Thinking": "Reasoning"
34
  },
35
  {
@@ -40,7 +590,7 @@
40
  "Result": 62.13,
41
  "Model URL": "https://huggingface.co/google",
42
  "Params": null,
43
- "Release": "V3",
44
  "Thinking": "Reasoning"
45
  },
46
  {
@@ -51,7 +601,7 @@
51
  "Result": 70.26,
52
  "Model URL": "https://huggingface.co/google",
53
  "Params": null,
54
- "Release": "V3",
55
  "Thinking": "Reasoning"
56
  },
57
  {
@@ -62,7 +612,7 @@
62
  "Result": 45.96,
63
  "Model URL": "https://huggingface.co/google",
64
  "Params": null,
65
- "Release": "V3",
66
  "Thinking": "Reasoning"
67
  },
68
  {
@@ -73,7 +623,7 @@
73
  "Result": 69.01,
74
  "Model URL": "https://huggingface.co/google",
75
  "Params": null,
76
- "Release": "V3",
77
  "Thinking": "Reasoning"
78
  },
79
  {
@@ -84,7 +634,7 @@
84
  "Result": 50.51,
85
  "Model URL": "https://huggingface.co/google",
86
  "Params": null,
87
- "Release": "V3",
88
  "Thinking": "Reasoning"
89
  },
90
  {
@@ -95,7 +645,7 @@
95
  "Result": 67.11,
96
  "Model URL": "https://huggingface.co/google",
97
  "Params": null,
98
- "Release": "V3",
99
  "Thinking": "Reasoning"
100
  },
101
  {
@@ -106,7 +656,7 @@
106
  "Result": 44.66,
107
  "Model URL": "https://huggingface.co/google",
108
  "Params": null,
109
- "Release": "V3",
110
  "Thinking": "Reasoning"
111
  },
112
  {
@@ -117,7 +667,7 @@
117
  "Result": 68.68,
118
  "Model URL": "https://huggingface.co/google",
119
  "Params": null,
120
- "Release": "V3",
121
  "Thinking": "Reasoning"
122
  },
123
  {
@@ -128,7 +678,7 @@
128
  "Result": 48.79,
129
  "Model URL": "https://huggingface.co/google",
130
  "Params": null,
131
- "Release": "V3",
132
  "Thinking": "Reasoning"
133
  },
134
  {
@@ -139,7 +689,7 @@
139
  "Result": -1.0,
140
  "Model URL": "https://huggingface.co/google",
141
  "Params": null,
142
- "Release": "V3",
143
  "Thinking": "Reasoning"
144
  },
145
  {
@@ -150,7 +700,7 @@
150
  "Result": 91.03,
151
  "Model URL": "https://huggingface.co/google",
152
  "Params": null,
153
- "Release": "V3",
154
  "Thinking": "Reasoning"
155
  },
156
  {
@@ -161,7 +711,7 @@
161
  "Result": 95.29,
162
  "Model URL": "https://huggingface.co/google",
163
  "Params": null,
164
- "Release": "V3",
165
  "Thinking": "Reasoning"
166
  },
167
  {
@@ -172,7 +722,7 @@
172
  "Result": 71.79,
173
  "Model URL": "https://huggingface.co/google",
174
  "Params": null,
175
- "Release": "V3",
176
  "Thinking": "Reasoning"
177
  },
178
  {
@@ -183,7 +733,7 @@
183
  "Result": 77.65,
184
  "Model URL": "https://huggingface.co/google",
185
  "Params": null,
186
- "Release": "V3",
187
  "Thinking": "Reasoning"
188
  },
189
  {
@@ -194,7 +744,7 @@
194
  "Result": 71.41,
195
  "Model URL": "https://huggingface.co/google",
196
  "Params": null,
197
- "Release": "V3",
198
  "Thinking": "Reasoning"
199
  },
200
  {
@@ -205,7 +755,7 @@
205
  "Result": 77.65,
206
  "Model URL": "https://huggingface.co/google",
207
  "Params": null,
208
- "Release": "V3",
209
  "Thinking": "Reasoning"
210
  },
211
  {
@@ -216,7 +766,7 @@
216
  "Result": 69.91,
217
  "Model URL": "https://huggingface.co/google",
218
  "Params": null,
219
- "Release": "V3",
220
  "Thinking": "Reasoning"
221
  },
222
  {
@@ -227,7 +777,7 @@
227
  "Result": 71.27,
228
  "Model URL": "https://huggingface.co/google",
229
  "Params": null,
230
- "Release": "V3",
231
  "Thinking": "Reasoning"
232
  },
233
  {
@@ -238,7 +788,7 @@
238
  "Result": 68.04,
239
  "Model URL": "https://huggingface.co/google",
240
  "Params": null,
241
- "Release": "V3",
242
  "Thinking": "Reasoning"
243
  },
244
  {
@@ -249,7 +799,7 @@
249
  "Result": 77.35,
250
  "Model URL": "https://huggingface.co/google",
251
  "Params": null,
252
- "Release": "V3",
253
  "Thinking": "Reasoning"
254
  },
255
  {
@@ -260,7 +810,7 @@
260
  "Result": 69.52,
261
  "Model URL": "https://huggingface.co/google",
262
  "Params": null,
263
- "Release": "V3",
264
  "Thinking": "Reasoning"
265
  },
266
  {
@@ -271,7 +821,7 @@
271
  "Result": 79.45,
272
  "Model URL": "https://huggingface.co/google",
273
  "Params": null,
274
- "Release": "V3",
275
  "Thinking": "Reasoning"
276
  },
277
  {
@@ -2482,7 +3032,7 @@
2482
  "Result": 88.97,
2483
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2484
  "Params": 36.2,
2485
- "Release": "V3",
2486
  "Thinking": "Reasoning"
2487
  },
2488
  {
@@ -2493,7 +3043,7 @@
2493
  "Result": 81.7,
2494
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2495
  "Params": 36.2,
2496
- "Release": "V3",
2497
  "Thinking": "Reasoning"
2498
  },
2499
  {
@@ -2504,7 +3054,7 @@
2504
  "Result": 71.15,
2505
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2506
  "Params": 36.2,
2507
- "Release": "V3",
2508
  "Thinking": "Reasoning"
2509
  },
2510
  {
@@ -2515,7 +3065,7 @@
2515
  "Result": 59.15,
2516
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2517
  "Params": 36.2,
2518
- "Release": "V3",
2519
  "Thinking": "Reasoning"
2520
  },
2521
  {
@@ -2526,7 +3076,7 @@
2526
  "Result": 71.15,
2527
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2528
  "Params": 36.2,
2529
- "Release": "V3",
2530
  "Thinking": "Reasoning"
2531
  },
2532
  {
@@ -2537,7 +3087,7 @@
2537
  "Result": 56.6,
2538
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2539
  "Params": 36.2,
2540
- "Release": "V3",
2541
  "Thinking": "Reasoning"
2542
  },
2543
  {
@@ -2548,7 +3098,7 @@
2548
  "Result": 70.1,
2549
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2550
  "Params": 36.2,
2551
- "Release": "V3",
2552
  "Thinking": "Reasoning"
2553
  },
2554
  {
@@ -2559,7 +3109,7 @@
2559
  "Result": 64.63,
2560
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2561
  "Params": 36.2,
2562
- "Release": "V3",
2563
  "Thinking": "Reasoning"
2564
  },
2565
  {
@@ -2570,7 +3120,7 @@
2570
  "Result": 68.1,
2571
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2572
  "Params": 36.2,
2573
- "Release": "V3",
2574
  "Thinking": "Reasoning"
2575
  },
2576
  {
@@ -2581,7 +3131,7 @@
2581
  "Result": 57.32,
2582
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2583
  "Params": 36.2,
2584
- "Release": "V3",
2585
  "Thinking": "Reasoning"
2586
  },
2587
  {
@@ -2592,7 +3142,7 @@
2592
  "Result": 70.32,
2593
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2594
  "Params": 36.2,
2595
- "Release": "V3",
2596
  "Thinking": "Reasoning"
2597
  },
2598
  {
@@ -2603,7 +3153,7 @@
2603
  "Result": 60.88,
2604
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2605
  "Params": 36.2,
2606
- "Release": "V3",
2607
  "Thinking": "Reasoning"
2608
  },
2609
  {
@@ -2614,7 +3164,7 @@
2614
  "Result": -1.0,
2615
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2616
  "Params": 36.2,
2617
- "Release": "V3",
2618
  "Thinking": "Reasoning"
2619
  },
2620
  {
@@ -2625,7 +3175,7 @@
2625
  "Result": 91.03,
2626
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2627
  "Params": 36.2,
2628
- "Release": "V3",
2629
  "Thinking": "Reasoning"
2630
  },
2631
  {
@@ -2636,7 +3186,7 @@
2636
  "Result": 91.76,
2637
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2638
  "Params": 36.2,
2639
- "Release": "V3",
2640
  "Thinking": "Reasoning"
2641
  },
2642
  {
@@ -2647,7 +3197,7 @@
2647
  "Result": 76.54,
2648
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2649
  "Params": 36.2,
2650
- "Release": "V3",
2651
  "Thinking": "Reasoning"
2652
  },
2653
  {
@@ -2658,7 +3208,7 @@
2658
  "Result": 70.59,
2659
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2660
  "Params": 36.2,
2661
- "Release": "V3",
2662
  "Thinking": "Reasoning"
2663
  },
2664
  {
@@ -2669,7 +3219,7 @@
2669
  "Result": 76.15,
2670
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2671
  "Params": 36.2,
2672
- "Release": "V3",
2673
  "Thinking": "Reasoning"
2674
  },
2675
  {
@@ -2680,7 +3230,7 @@
2680
  "Result": 70.59,
2681
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2682
  "Params": 36.2,
2683
- "Release": "V3",
2684
  "Thinking": "Reasoning"
2685
  },
2686
  {
@@ -2691,7 +3241,7 @@
2691
  "Result": 74.95,
2692
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2693
  "Params": 36.2,
2694
- "Release": "V3",
2695
  "Thinking": "Reasoning"
2696
  },
2697
  {
@@ -2702,7 +3252,7 @@
2702
  "Result": 66.11,
2703
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2704
  "Params": 36.2,
2705
- "Release": "V3",
2706
  "Thinking": "Reasoning"
2707
  },
2708
  {
@@ -2713,7 +3263,7 @@
2713
  "Result": 73.33,
2714
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2715
  "Params": 36.2,
2716
- "Release": "V3",
2717
  "Thinking": "Reasoning"
2718
  },
2719
  {
@@ -2724,7 +3274,7 @@
2724
  "Result": 70.19,
2725
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2726
  "Params": 36.2,
2727
- "Release": "V3",
2728
  "Thinking": "Reasoning"
2729
  },
2730
  {
@@ -2735,7 +3285,7 @@
2735
  "Result": 74.69,
2736
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2737
  "Params": 36.2,
2738
- "Release": "V3",
2739
  "Thinking": "Reasoning"
2740
  },
2741
  {
@@ -2746,7 +3296,7 @@
2746
  "Result": 72.45,
2747
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2748
  "Params": 36.2,
2749
- "Release": "V3",
2750
  "Thinking": "Reasoning"
2751
  },
2752
  {
@@ -4407,7 +4957,7 @@
4407
  "Result": 79.74,
4408
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4409
  "Params": 14,
4410
- "Release": "V3",
4411
  "Thinking": "Reasoning"
4412
  },
4413
  {
@@ -4418,7 +4968,7 @@
4418
  "Result": 76.6,
4419
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4420
  "Params": 14,
4421
- "Release": "V3",
4422
  "Thinking": "Reasoning"
4423
  },
4424
  {
@@ -4429,7 +4979,7 @@
4429
  "Result": 55.9,
4430
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4431
  "Params": 14,
4432
- "Release": "V3",
4433
  "Thinking": "Reasoning"
4434
  },
4435
  {
@@ -4440,7 +4990,7 @@
4440
  "Result": 39.57,
4441
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4442
  "Params": 14,
4443
- "Release": "V3",
4444
  "Thinking": "Reasoning"
4445
  },
4446
  {
@@ -4451,7 +5001,7 @@
4451
  "Result": 55.64,
4452
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4453
  "Params": 14,
4454
- "Release": "V3",
4455
  "Thinking": "Reasoning"
4456
  },
4457
  {
@@ -4462,7 +5012,7 @@
4462
  "Result": 34.47,
4463
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4464
  "Params": 14,
4465
- "Release": "V3",
4466
  "Thinking": "Reasoning"
4467
  },
4468
  {
@@ -4473,7 +5023,7 @@
4473
  "Result": 55.34,
4474
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4475
  "Params": 14,
4476
- "Release": "V3",
4477
  "Thinking": "Reasoning"
4478
  },
4479
  {
@@ -4484,7 +5034,7 @@
4484
  "Result": 38.35,
4485
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4486
  "Params": 14,
4487
- "Release": "V3",
4488
  "Thinking": "Reasoning"
4489
  },
4490
  {
@@ -4495,7 +5045,7 @@
4495
  "Result": 53.73,
4496
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4497
  "Params": 14,
4498
- "Release": "V3",
4499
  "Thinking": "Reasoning"
4500
  },
4501
  {
@@ -4506,7 +5056,7 @@
4506
  "Result": 32.73,
4507
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4508
  "Params": 14,
4509
- "Release": "V3",
4510
  "Thinking": "Reasoning"
4511
  },
4512
  {
@@ -4517,7 +5067,7 @@
4517
  "Result": 55.17,
4518
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4519
  "Params": 14,
4520
- "Release": "V3",
4521
  "Thinking": "Reasoning"
4522
  },
4523
  {
@@ -4528,7 +5078,7 @@
4528
  "Result": 36.13,
4529
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4530
  "Params": 14,
4531
- "Release": "V3",
4532
  "Thinking": "Reasoning"
4533
  },
4534
  {
@@ -4539,7 +5089,7 @@
4539
  "Result": -1.0,
4540
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4541
  "Params": 14,
4542
- "Release": "V3",
4543
  "Thinking": "Reasoning"
4544
  },
4545
  {
@@ -4550,7 +5100,7 @@
4550
  "Result": 51.15,
4551
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4552
  "Params": 14,
4553
- "Release": "V3",
4554
  "Thinking": "Reasoning"
4555
  },
4556
  {
@@ -4561,7 +5111,7 @@
4561
  "Result": 74.12,
4562
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4563
  "Params": 14,
4564
- "Release": "V3",
4565
  "Thinking": "Reasoning"
4566
  },
4567
  {
@@ -4572,7 +5122,7 @@
4572
  "Result": 36.54,
4573
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4574
  "Params": 14,
4575
- "Release": "V3",
4576
  "Thinking": "Reasoning"
4577
  },
4578
  {
@@ -4583,7 +5133,7 @@
4583
  "Result": 56.47,
4584
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4585
  "Params": 14,
4586
- "Release": "V3",
4587
  "Thinking": "Reasoning"
4588
  },
4589
  {
@@ -4594,7 +5144,7 @@
4594
  "Result": 36.28,
4595
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4596
  "Params": 14,
4597
- "Release": "V3",
4598
  "Thinking": "Reasoning"
4599
  },
4600
  {
@@ -4605,7 +5155,7 @@
4605
  "Result": 56.47,
4606
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4607
  "Params": 14,
4608
- "Release": "V3",
4609
  "Thinking": "Reasoning"
4610
  },
4611
  {
@@ -4616,7 +5166,7 @@
4616
  "Result": 35.26,
4617
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4618
  "Params": 14,
4619
- "Release": "V3",
4620
  "Thinking": "Reasoning"
4621
  },
4622
  {
@@ -4627,7 +5177,7 @@
4627
  "Result": 52.97,
4628
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4629
  "Params": 14,
4630
- "Release": "V3",
4631
  "Thinking": "Reasoning"
4632
  },
4633
  {
@@ -4638,7 +5188,7 @@
4638
  "Result": 35.52,
4639
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4640
  "Params": 14,
4641
- "Release": "V3",
4642
  "Thinking": "Reasoning"
4643
  },
4644
  {
@@ -4649,7 +5199,7 @@
4649
  "Result": 56.22,
4650
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4651
  "Params": 14,
4652
- "Release": "V3",
4653
  "Thinking": "Reasoning"
4654
  },
4655
  {
@@ -4660,7 +5210,7 @@
4660
  "Result": 34.94,
4661
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4662
  "Params": 14,
4663
- "Release": "V3",
4664
  "Thinking": "Reasoning"
4665
  },
4666
  {
@@ -4671,7 +5221,7 @@
4671
  "Result": 57.44,
4672
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4673
  "Params": 14,
4674
- "Release": "V3",
4675
  "Thinking": "Reasoning"
4676
  },
4677
  {
@@ -4682,7 +5232,7 @@
4682
  "Result": 72.18,
4683
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4684
  "Params": 14,
4685
- "Release": "V3",
4686
  "Thinking": "Dense"
4687
  },
4688
  {
@@ -4693,7 +5243,7 @@
4693
  "Result": 80.85,
4694
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4695
  "Params": 14,
4696
- "Release": "V3",
4697
  "Thinking": "Dense"
4698
  },
4699
  {
@@ -4704,7 +5254,7 @@
4704
  "Result": 42.31,
4705
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4706
  "Params": 14,
4707
- "Release": "V3",
4708
  "Thinking": "Dense"
4709
  },
4710
  {
@@ -4715,7 +5265,7 @@
4715
  "Result": 48.09,
4716
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4717
  "Params": 14,
4718
- "Release": "V3",
4719
  "Thinking": "Dense"
4720
  },
4721
  {
@@ -4726,7 +5276,7 @@
4726
  "Result": 42.31,
4727
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4728
  "Params": 14,
4729
- "Release": "V3",
4730
  "Thinking": "Dense"
4731
  },
4732
  {
@@ -4737,7 +5287,7 @@
4737
  "Result": 45.53,
4738
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4739
  "Params": 14,
4740
- "Release": "V3",
4741
  "Thinking": "Dense"
4742
  },
4743
  {
@@ -4748,7 +5298,7 @@
4748
  "Result": 41.73,
4749
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4750
  "Params": 14,
4751
- "Release": "V3",
4752
  "Thinking": "Dense"
4753
  },
4754
  {
@@ -4759,7 +5309,7 @@
4759
  "Result": 50.59,
4760
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4761
  "Params": 14,
4762
- "Release": "V3",
4763
  "Thinking": "Dense"
4764
  },
4765
  {
@@ -4770,7 +5320,7 @@
4770
  "Result": 40.33,
4771
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4772
  "Params": 14,
4773
- "Release": "V3",
4774
  "Thinking": "Dense"
4775
  },
4776
  {
@@ -4781,7 +5331,7 @@
4781
  "Result": 45.81,
4782
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4783
  "Params": 14,
4784
- "Release": "V3",
4785
  "Thinking": "Dense"
4786
  },
4787
  {
@@ -4792,7 +5342,7 @@
4792
  "Result": 41.38,
4793
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4794
  "Params": 14,
4795
- "Release": "V3",
4796
  "Thinking": "Dense"
4797
  },
4798
  {
@@ -4803,7 +5353,7 @@
4803
  "Result": 47.91,
4804
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4805
  "Params": 14,
4806
- "Release": "V3",
4807
  "Thinking": "Dense"
4808
  },
4809
  {
@@ -4814,7 +5364,7 @@
4814
  "Result": 28.36,
4815
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4816
  "Params": 14,
4817
- "Release": "V3",
4818
  "Thinking": "Dense"
4819
  },
4820
  {
@@ -4825,7 +5375,7 @@
4825
  "Result": 52.69,
4826
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4827
  "Params": 14,
4828
- "Release": "V3",
4829
  "Thinking": "Dense"
4830
  },
4831
  {
@@ -4836,7 +5386,7 @@
4836
  "Result": 83.53,
4837
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4838
  "Params": 14,
4839
- "Release": "V3",
4840
  "Thinking": "Dense"
4841
  },
4842
  {
@@ -4847,7 +5397,7 @@
4847
  "Result": 27.44,
4848
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4849
  "Params": 14,
4850
- "Release": "V3",
4851
  "Thinking": "Dense"
4852
  },
4853
  {
@@ -4858,7 +5408,7 @@
4858
  "Result": 45.88,
4859
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4860
  "Params": 14,
4861
- "Release": "V3",
4862
  "Thinking": "Dense"
4863
  },
4864
  {
@@ -4869,7 +5419,7 @@
4869
  "Result": 27.44,
4870
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4871
  "Params": 14,
4872
- "Release": "V3",
4873
  "Thinking": "Dense"
4874
  },
4875
  {
@@ -4880,7 +5430,7 @@
4880
  "Result": 45.88,
4881
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4882
  "Params": 14,
4883
- "Release": "V3",
4884
  "Thinking": "Dense"
4885
  },
4886
  {
@@ -4891,7 +5441,7 @@
4891
  "Result": 26.54,
4892
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4893
  "Params": 14,
4894
- "Release": "V3",
4895
  "Thinking": "Dense"
4896
  },
4897
  {
@@ -4902,7 +5452,7 @@
4902
  "Result": 43.55,
4903
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4904
  "Params": 14,
4905
- "Release": "V3",
4906
  "Thinking": "Dense"
4907
  },
4908
  {
@@ -4913,7 +5463,7 @@
4913
  "Result": 26.68,
4914
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4915
  "Params": 14,
4916
- "Release": "V3",
4917
  "Thinking": "Dense"
4918
  },
4919
  {
@@ -4924,7 +5474,7 @@
4924
  "Result": 45.98,
4925
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4926
  "Params": 14,
4927
- "Release": "V3",
4928
  "Thinking": "Dense"
4929
  },
4930
  {
@@ -4935,7 +5485,7 @@
4935
  "Result": 25.11,
4936
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4937
  "Params": 14,
4938
- "Release": "V3",
4939
  "Thinking": "Dense"
4940
  },
4941
  {
@@ -4946,7 +5496,7 @@
4946
  "Result": 44.96,
4947
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4948
  "Params": 14,
4949
- "Release": "V3",
4950
  "Thinking": "Dense"
4951
  },
4952
  {
@@ -4957,7 +5507,7 @@
4957
  "Result": 70.77,
4958
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4959
  "Params": 8.2,
4960
- "Release": "V3",
4961
  "Thinking": "Reasoning"
4962
  },
4963
  {
@@ -4968,7 +5518,7 @@
4968
  "Result": 62.13,
4969
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4970
  "Params": 8.2,
4971
- "Release": "V3",
4972
  "Thinking": "Reasoning"
4973
  },
4974
  {
@@ -4979,7 +5529,7 @@
4979
  "Result": 50.0,
4980
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4981
  "Params": 8.2,
4982
- "Release": "V3",
4983
  "Thinking": "Reasoning"
4984
  },
4985
  {
@@ -4990,7 +5540,7 @@
4990
  "Result": 34.47,
4991
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4992
  "Params": 8.2,
4993
- "Release": "V3",
4994
  "Thinking": "Reasoning"
4995
  },
4996
  {
@@ -5001,7 +5551,7 @@
5001
  "Result": 50.0,
5002
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5003
  "Params": 8.2,
5004
- "Release": "V3",
5005
  "Thinking": "Reasoning"
5006
  },
5007
  {
@@ -5012,7 +5562,7 @@
5012
  "Result": 32.77,
5013
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5014
  "Params": 8.2,
5015
- "Release": "V3",
5016
  "Thinking": "Reasoning"
5017
  },
5018
  {
@@ -5023,7 +5573,7 @@
5023
  "Result": 49.33,
5024
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5025
  "Params": 8.2,
5026
- "Release": "V3",
5027
  "Thinking": "Reasoning"
5028
  },
5029
  {
@@ -5034,7 +5584,7 @@
5034
  "Result": 35.1,
5035
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5036
  "Params": 8.2,
5037
- "Release": "V3",
5038
  "Thinking": "Reasoning"
5039
  },
5040
  {
@@ -5045,7 +5595,7 @@
5045
  "Result": 47.92,
5046
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5047
  "Params": 8.2,
5048
- "Release": "V3",
5049
  "Thinking": "Reasoning"
5050
  },
5051
  {
@@ -5056,7 +5606,7 @@
5056
  "Result": 32.52,
5057
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5058
  "Params": 8.2,
5059
- "Release": "V3",
5060
  "Thinking": "Reasoning"
5061
  },
5062
  {
@@ -5067,7 +5617,7 @@
5067
  "Result": 48.91,
5068
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5069
  "Params": 8.2,
5070
- "Release": "V3",
5071
  "Thinking": "Reasoning"
5072
  },
5073
  {
@@ -5078,7 +5628,7 @@
5078
  "Result": 31.88,
5079
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5080
  "Params": 8.2,
5081
- "Release": "V3",
5082
  "Thinking": "Reasoning"
5083
  },
5084
  {
@@ -5089,7 +5639,7 @@
5089
  "Result": -1.0,
5090
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5091
  "Params": 8.2,
5092
- "Release": "V3",
5093
  "Thinking": "Reasoning"
5094
  },
5095
  {
@@ -5100,7 +5650,7 @@
5100
  "Result": 69.36,
5101
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5102
  "Params": 8.2,
5103
- "Release": "V3",
5104
  "Thinking": "Reasoning"
5105
  },
5106
  {
@@ -5111,7 +5661,7 @@
5111
  "Result": 81.18,
5112
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5113
  "Params": 8.2,
5114
- "Release": "V3",
5115
  "Thinking": "Reasoning"
5116
  },
5117
  {
@@ -5122,7 +5672,7 @@
5122
  "Result": 51.54,
5123
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5124
  "Params": 8.2,
5125
- "Release": "V3",
5126
  "Thinking": "Reasoning"
5127
  },
5128
  {
@@ -5133,7 +5683,7 @@
5133
  "Result": 40.0,
5134
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5135
  "Params": 8.2,
5136
- "Release": "V3",
5137
  "Thinking": "Reasoning"
5138
  },
5139
  {
@@ -5144,7 +5694,7 @@
5144
  "Result": 50.64,
5145
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5146
  "Params": 8.2,
5147
- "Release": "V3",
5148
  "Thinking": "Reasoning"
5149
  },
5150
  {
@@ -5155,7 +5705,7 @@
5155
  "Result": 38.82,
5156
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5157
  "Params": 8.2,
5158
- "Release": "V3",
5159
  "Thinking": "Reasoning"
5160
  },
5161
  {
@@ -5166,7 +5716,7 @@
5166
  "Result": 49.91,
5167
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5168
  "Params": 8.2,
5169
- "Release": "V3",
5170
  "Thinking": "Reasoning"
5171
  },
5172
  {
@@ -5177,7 +5727,7 @@
5177
  "Result": 37.47,
5178
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5179
  "Params": 8.2,
5180
- "Release": "V3",
5181
  "Thinking": "Reasoning"
5182
  },
5183
  {
@@ -5188,7 +5738,7 @@
5188
  "Result": 48.4,
5189
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5190
  "Params": 8.2,
5191
- "Release": "V3",
5192
  "Thinking": "Reasoning"
5193
  },
5194
  {
@@ -5199,7 +5749,7 @@
5199
  "Result": 38.76,
5200
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5201
  "Params": 8.2,
5202
- "Release": "V3",
5203
  "Thinking": "Reasoning"
5204
  },
5205
  {
@@ -5210,7 +5760,7 @@
5210
  "Result": 49.41,
5211
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5212
  "Params": 8.2,
5213
- "Release": "V3",
5214
  "Thinking": "Reasoning"
5215
  },
5216
  {
@@ -5221,7 +5771,7 @@
5221
  "Result": 38.73,
5222
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5223
  "Params": 8.2,
5224
- "Release": "V3",
5225
  "Thinking": "Reasoning"
5226
  },
5227
  {
 
1
  [
2
+ {
3
+ "Model": "DeepSeek V3.1 Terminus",
4
+ "Model Type": "General",
5
+ "Benchmark": "VerilogEval S2R",
6
+ "Task": "Syntax (STX)",
7
+ "Result": 96.28,
8
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
9
+ "Params": 685,
10
+ "Release": "V4",
11
+ "Thinking": "Reasoning"
12
+ },
13
+ {
14
+ "Model": "DeepSeek V3.1 Terminus",
15
+ "Model Type": "General",
16
+ "Benchmark": "RTLLM",
17
+ "Task": "Syntax (STX)",
18
+ "Result": 90.21,
19
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
20
+ "Params": 685,
21
+ "Release": "V4",
22
+ "Thinking": "Reasoning"
23
+ },
24
+ {
25
+ "Model": "DeepSeek V3.1 Terminus",
26
+ "Model Type": "General",
27
+ "Benchmark": "VerilogEval S2R",
28
+ "Task": "Functionality (FNC)",
29
+ "Result": 76.79,
30
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
31
+ "Params": 685,
32
+ "Release": "V4",
33
+ "Thinking": "Reasoning"
34
+ },
35
+ {
36
+ "Model": "DeepSeek V3.1 Terminus",
37
+ "Model Type": "General",
38
+ "Benchmark": "RTLLM",
39
+ "Task": "Functionality (FNC)",
40
+ "Result": 60.43,
41
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
42
+ "Params": 685,
43
+ "Release": "V4",
44
+ "Thinking": "Reasoning"
45
+ },
46
+ {
47
+ "Model": "DeepSeek V3.1 Terminus",
48
+ "Model Type": "General",
49
+ "Benchmark": "VerilogEval S2R",
50
+ "Task": "Synthesis (SYN)",
51
+ "Result": 76.67,
52
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
53
+ "Params": 685,
54
+ "Release": "V4",
55
+ "Thinking": "Reasoning"
56
+ },
57
+ {
58
+ "Model": "DeepSeek V3.1 Terminus",
59
+ "Model Type": "General",
60
+ "Benchmark": "RTLLM",
61
+ "Task": "Synthesis (SYN)",
62
+ "Result": 58.3,
63
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
64
+ "Params": 685,
65
+ "Release": "V4",
66
+ "Thinking": "Reasoning"
67
+ },
68
+ {
69
+ "Model": "DeepSeek V3.1 Terminus",
70
+ "Model Type": "General",
71
+ "Benchmark": "VerilogEval S2R",
72
+ "Task": "Power",
73
+ "Result": 76.05,
74
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
75
+ "Params": 685,
76
+ "Release": "V4",
77
+ "Thinking": "Reasoning"
78
+ },
79
+ {
80
+ "Model": "DeepSeek V3.1 Terminus",
81
+ "Model Type": "General",
82
+ "Benchmark": "RTLLM",
83
+ "Task": "Power",
84
+ "Result": 63.84,
85
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
86
+ "Params": 685,
87
+ "Release": "V4",
88
+ "Thinking": "Reasoning"
89
+ },
90
+ {
91
+ "Model": "DeepSeek V3.1 Terminus",
92
+ "Model Type": "General",
93
+ "Benchmark": "VerilogEval S2R",
94
+ "Task": "Performance",
95
+ "Result": 73.76,
96
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
97
+ "Params": 685,
98
+ "Release": "V4",
99
+ "Thinking": "Reasoning"
100
+ },
101
+ {
102
+ "Model": "DeepSeek V3.1 Terminus",
103
+ "Model Type": "General",
104
+ "Benchmark": "RTLLM",
105
+ "Task": "Performance",
106
+ "Result": 58.77,
107
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
108
+ "Params": 685,
109
+ "Release": "V4",
110
+ "Thinking": "Reasoning"
111
+ },
112
+ {
113
+ "Model": "DeepSeek V3.1 Terminus",
114
+ "Model Type": "General",
115
+ "Benchmark": "VerilogEval S2R",
116
+ "Task": "Area",
117
+ "Result": 74.94,
118
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
119
+ "Params": 685,
120
+ "Release": "V4",
121
+ "Thinking": "Reasoning"
122
+ },
123
+ {
124
+ "Model": "DeepSeek V3.1 Terminus",
125
+ "Model Type": "General",
126
+ "Benchmark": "RTLLM",
127
+ "Task": "Area",
128
+ "Result": 61.83,
129
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
130
+ "Params": 685,
131
+ "Release": "V4",
132
+ "Thinking": "Reasoning"
133
+ },
134
+ {
135
+ "Model": "DeepSeek V3.1 Terminus",
136
+ "Model Type": "General",
137
+ "Benchmark": "RTL-Repo",
138
+ "Task": "Exact Matching (EM)",
139
+ "Result": -1.0,
140
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
141
+ "Params": 685,
142
+ "Release": "V4",
143
+ "Thinking": "Reasoning"
144
+ },
145
+ {
146
+ "Model": "DeepSeek V3.1 Terminus",
147
+ "Model Type": "General",
148
+ "Benchmark": "VerilogEval MC",
149
+ "Task": "Syntax (STX)",
150
+ "Result": 95.38,
151
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
152
+ "Params": 685,
153
+ "Release": "V4",
154
+ "Thinking": "Reasoning"
155
+ },
156
+ {
157
+ "Model": "DeepSeek V3.1 Terminus",
158
+ "Model Type": "General",
159
+ "Benchmark": "VeriGen",
160
+ "Task": "Syntax (STX)",
161
+ "Result": 92.94,
162
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
163
+ "Params": 685,
164
+ "Release": "V4",
165
+ "Thinking": "Reasoning"
166
+ },
167
+ {
168
+ "Model": "DeepSeek V3.1 Terminus",
169
+ "Model Type": "General",
170
+ "Benchmark": "VerilogEval MC",
171
+ "Task": "Functionality (FNC)",
172
+ "Result": 78.97,
173
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
174
+ "Params": 685,
175
+ "Release": "V4",
176
+ "Thinking": "Reasoning"
177
+ },
178
+ {
179
+ "Model": "DeepSeek V3.1 Terminus",
180
+ "Model Type": "General",
181
+ "Benchmark": "VeriGen",
182
+ "Task": "Functionality (FNC)",
183
+ "Result": 75.29,
184
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
185
+ "Params": 685,
186
+ "Release": "V4",
187
+ "Thinking": "Reasoning"
188
+ },
189
+ {
190
+ "Model": "DeepSeek V3.1 Terminus",
191
+ "Model Type": "General",
192
+ "Benchmark": "VerilogEval MC",
193
+ "Task": "Synthesis (SYN)",
194
+ "Result": 78.85,
195
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
196
+ "Params": 685,
197
+ "Release": "V4",
198
+ "Thinking": "Reasoning"
199
+ },
200
+ {
201
+ "Model": "DeepSeek V3.1 Terminus",
202
+ "Model Type": "General",
203
+ "Benchmark": "VeriGen",
204
+ "Task": "Synthesis (SYN)",
205
+ "Result": 75.29,
206
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
207
+ "Params": 685,
208
+ "Release": "V4",
209
+ "Thinking": "Reasoning"
210
+ },
211
+ {
212
+ "Model": "DeepSeek V3.1 Terminus",
213
+ "Model Type": "General",
214
+ "Benchmark": "VerilogEval MC",
215
+ "Task": "Power",
216
+ "Result": 77.34,
217
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
218
+ "Params": 685,
219
+ "Release": "V4",
220
+ "Thinking": "Reasoning"
221
+ },
222
+ {
223
+ "Model": "DeepSeek V3.1 Terminus",
224
+ "Model Type": "General",
225
+ "Benchmark": "VeriGen",
226
+ "Task": "Power",
227
+ "Result": 69.87,
228
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
229
+ "Params": 685,
230
+ "Release": "V4",
231
+ "Thinking": "Reasoning"
232
+ },
233
+ {
234
+ "Model": "DeepSeek V3.1 Terminus",
235
+ "Model Type": "General",
236
+ "Benchmark": "VerilogEval MC",
237
+ "Task": "Performance",
238
+ "Result": 76.11,
239
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
240
+ "Params": 685,
241
+ "Release": "V4",
242
+ "Thinking": "Reasoning"
243
+ },
244
+ {
245
+ "Model": "DeepSeek V3.1 Terminus",
246
+ "Model Type": "General",
247
+ "Benchmark": "VeriGen",
248
+ "Task": "Performance",
249
+ "Result": 75.14,
250
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
251
+ "Params": 685,
252
+ "Release": "V4",
253
+ "Thinking": "Reasoning"
254
+ },
255
+ {
256
+ "Model": "DeepSeek V3.1 Terminus",
257
+ "Model Type": "General",
258
+ "Benchmark": "VerilogEval MC",
259
+ "Task": "Area",
260
+ "Result": 77.03,
261
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
262
+ "Params": 685,
263
+ "Release": "V4",
264
+ "Thinking": "Reasoning"
265
+ },
266
+ {
267
+ "Model": "DeepSeek V3.1 Terminus",
268
+ "Model Type": "General",
269
+ "Benchmark": "VeriGen",
270
+ "Task": "Area",
271
+ "Result": 77.59,
272
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
273
+ "Params": 685,
274
+ "Release": "V4",
275
+ "Thinking": "Reasoning"
276
+ },
277
+ {
278
+ "Model": "Kimi K2 Instruct 0905",
279
+ "Model Type": "General",
280
+ "Benchmark": "VerilogEval S2R",
281
+ "Task": "Syntax (STX)",
282
+ "Result": 96.15,
283
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
284
+ "Params": 1000,
285
+ "Release": "V4",
286
+ "Thinking": "Dense"
287
+ },
288
+ {
289
+ "Model": "Kimi K2 Instruct 0905",
290
+ "Model Type": "General",
291
+ "Benchmark": "RTLLM",
292
+ "Task": "Syntax (STX)",
293
+ "Result": 87.23,
294
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
295
+ "Params": 1000,
296
+ "Release": "V4",
297
+ "Thinking": "Dense"
298
+ },
299
+ {
300
+ "Model": "Kimi K2 Instruct 0905",
301
+ "Model Type": "General",
302
+ "Benchmark": "VerilogEval S2R",
303
+ "Task": "Functionality (FNC)",
304
+ "Result": 72.82,
305
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
306
+ "Params": 1000,
307
+ "Release": "V4",
308
+ "Thinking": "Dense"
309
+ },
310
+ {
311
+ "Model": "Kimi K2 Instruct 0905",
312
+ "Model Type": "General",
313
+ "Benchmark": "RTLLM",
314
+ "Task": "Functionality (FNC)",
315
+ "Result": 62.13,
316
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
317
+ "Params": 1000,
318
+ "Release": "V4",
319
+ "Thinking": "Dense"
320
+ },
321
+ {
322
+ "Model": "Kimi K2 Instruct 0905",
323
+ "Model Type": "General",
324
+ "Benchmark": "VerilogEval S2R",
325
+ "Task": "Synthesis (SYN)",
326
+ "Result": 72.69,
327
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
328
+ "Params": 1000,
329
+ "Release": "V4",
330
+ "Thinking": "Dense"
331
+ },
332
+ {
333
+ "Model": "Kimi K2 Instruct 0905",
334
+ "Model Type": "General",
335
+ "Benchmark": "RTLLM",
336
+ "Task": "Synthesis (SYN)",
337
+ "Result": 60.43,
338
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
339
+ "Params": 1000,
340
+ "Release": "V4",
341
+ "Thinking": "Dense"
342
+ },
343
+ {
344
+ "Model": "Kimi K2 Instruct 0905",
345
+ "Model Type": "General",
346
+ "Benchmark": "VerilogEval S2R",
347
+ "Task": "Power",
348
+ "Result": 71.28,
349
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
350
+ "Params": 1000,
351
+ "Release": "V4",
352
+ "Thinking": "Dense"
353
+ },
354
+ {
355
+ "Model": "Kimi K2 Instruct 0905",
356
+ "Model Type": "General",
357
+ "Benchmark": "RTLLM",
358
+ "Task": "Power",
359
+ "Result": 63.48,
360
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
361
+ "Params": 1000,
362
+ "Release": "V4",
363
+ "Thinking": "Dense"
364
+ },
365
+ {
366
+ "Model": "Kimi K2 Instruct 0905",
367
+ "Model Type": "General",
368
+ "Benchmark": "VerilogEval S2R",
369
+ "Task": "Performance",
370
+ "Result": 69.98,
371
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
372
+ "Params": 1000,
373
+ "Release": "V4",
374
+ "Thinking": "Dense"
375
+ },
376
+ {
377
+ "Model": "Kimi K2 Instruct 0905",
378
+ "Model Type": "General",
379
+ "Benchmark": "RTLLM",
380
+ "Task": "Performance",
381
+ "Result": 58.58,
382
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
383
+ "Params": 1000,
384
+ "Release": "V4",
385
+ "Thinking": "Dense"
386
+ },
387
+ {
388
+ "Model": "Kimi K2 Instruct 0905",
389
+ "Model Type": "General",
390
+ "Benchmark": "VerilogEval S2R",
391
+ "Task": "Area",
392
+ "Result": 71.2,
393
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
394
+ "Params": 1000,
395
+ "Release": "V4",
396
+ "Thinking": "Dense"
397
+ },
398
+ {
399
+ "Model": "Kimi K2 Instruct 0905",
400
+ "Model Type": "General",
401
+ "Benchmark": "RTLLM",
402
+ "Task": "Area",
403
+ "Result": 63.32,
404
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
405
+ "Params": 1000,
406
+ "Release": "V4",
407
+ "Thinking": "Dense"
408
+ },
409
+ {
410
+ "Model": "Kimi K2 Instruct 0905",
411
+ "Model Type": "General",
412
+ "Benchmark": "RTL-Repo",
413
+ "Task": "Exact Matching (EM)",
414
+ "Result": 33.65,
415
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
416
+ "Params": 1000,
417
+ "Release": "V4",
418
+ "Thinking": "Dense"
419
+ },
420
+ {
421
+ "Model": "Kimi K2 Instruct 0905",
422
+ "Model Type": "General",
423
+ "Benchmark": "VerilogEval MC",
424
+ "Task": "Syntax (STX)",
425
+ "Result": 98.21,
426
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
427
+ "Params": 1000,
428
+ "Release": "V4",
429
+ "Thinking": "Dense"
430
+ },
431
+ {
432
+ "Model": "Kimi K2 Instruct 0905",
433
+ "Model Type": "General",
434
+ "Benchmark": "VeriGen",
435
+ "Task": "Syntax (STX)",
436
+ "Result": 95.29,
437
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
438
+ "Params": 1000,
439
+ "Release": "V4",
440
+ "Thinking": "Dense"
441
+ },
442
+ {
443
+ "Model": "Kimi K2 Instruct 0905",
444
+ "Model Type": "General",
445
+ "Benchmark": "VerilogEval MC",
446
+ "Task": "Functionality (FNC)",
447
+ "Result": 74.74,
448
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
449
+ "Params": 1000,
450
+ "Release": "V4",
451
+ "Thinking": "Dense"
452
+ },
453
+ {
454
+ "Model": "Kimi K2 Instruct 0905",
455
+ "Model Type": "General",
456
+ "Benchmark": "VeriGen",
457
+ "Task": "Functionality (FNC)",
458
+ "Result": 69.41,
459
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
460
+ "Params": 1000,
461
+ "Release": "V4",
462
+ "Thinking": "Dense"
463
+ },
464
+ {
465
+ "Model": "Kimi K2 Instruct 0905",
466
+ "Model Type": "General",
467
+ "Benchmark": "VerilogEval MC",
468
+ "Task": "Synthesis (SYN)",
469
+ "Result": 74.49,
470
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
471
+ "Params": 1000,
472
+ "Release": "V4",
473
+ "Thinking": "Dense"
474
+ },
475
+ {
476
+ "Model": "Kimi K2 Instruct 0905",
477
+ "Model Type": "General",
478
+ "Benchmark": "VeriGen",
479
+ "Task": "Synthesis (SYN)",
480
+ "Result": 69.41,
481
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
482
+ "Params": 1000,
483
+ "Release": "V4",
484
+ "Thinking": "Dense"
485
+ },
486
+ {
487
+ "Model": "Kimi K2 Instruct 0905",
488
+ "Model Type": "General",
489
+ "Benchmark": "VerilogEval MC",
490
+ "Task": "Power",
491
+ "Result": 73.13,
492
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
493
+ "Params": 1000,
494
+ "Release": "V4",
495
+ "Thinking": "Dense"
496
+ },
497
+ {
498
+ "Model": "Kimi K2 Instruct 0905",
499
+ "Model Type": "General",
500
+ "Benchmark": "VeriGen",
501
+ "Task": "Power",
502
+ "Result": 57.47,
503
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
504
+ "Params": 1000,
505
+ "Release": "V4",
506
+ "Thinking": "Dense"
507
+ },
508
+ {
509
+ "Model": "Kimi K2 Instruct 0905",
510
+ "Model Type": "General",
511
+ "Benchmark": "VerilogEval MC",
512
+ "Task": "Performance",
513
+ "Result": 71.43,
514
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
515
+ "Params": 1000,
516
+ "Release": "V4",
517
+ "Thinking": "Dense"
518
+ },
519
+ {
520
+ "Model": "Kimi K2 Instruct 0905",
521
+ "Model Type": "General",
522
+ "Benchmark": "VeriGen",
523
+ "Task": "Performance",
524
+ "Result": 69.9,
525
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
526
+ "Params": 1000,
527
+ "Release": "V4",
528
+ "Thinking": "Dense"
529
+ },
530
+ {
531
+ "Model": "Kimi K2 Instruct 0905",
532
+ "Model Type": "General",
533
+ "Benchmark": "VerilogEval MC",
534
+ "Task": "Area",
535
+ "Result": 72.76,
536
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
537
+ "Params": 1000,
538
+ "Release": "V4",
539
+ "Thinking": "Dense"
540
+ },
541
+ {
542
+ "Model": "Kimi K2 Instruct 0905",
543
+ "Model Type": "General",
544
+ "Benchmark": "VeriGen",
545
+ "Task": "Area",
546
+ "Result": 69.51,
547
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
548
+ "Params": 1000,
549
+ "Release": "V4",
550
+ "Thinking": "Dense"
551
+ },
552
  {
553
  "Model": "Gemini 2.5 Flash (Medium)",
554
  "Model Type": "General",
 
557
  "Result": 92.05,
558
  "Model URL": "https://huggingface.co/google",
559
  "Params": null,
560
+ "Release": "V4",
561
  "Thinking": "Reasoning"
562
  },
563
  {
 
568
  "Result": 79.57,
569
  "Model URL": "https://huggingface.co/google",
570
  "Params": null,
571
+ "Release": "V4",
572
  "Thinking": "Reasoning"
573
  },
574
  {
 
579
  "Result": 70.38,
580
  "Model URL": "https://huggingface.co/google",
581
  "Params": null,
582
+ "Release": "V4",
583
  "Thinking": "Reasoning"
584
  },
585
  {
 
590
  "Result": 62.13,
591
  "Model URL": "https://huggingface.co/google",
592
  "Params": null,
593
+ "Release": "V4",
594
  "Thinking": "Reasoning"
595
  },
596
  {
 
601
  "Result": 70.26,
602
  "Model URL": "https://huggingface.co/google",
603
  "Params": null,
604
+ "Release": "V4",
605
  "Thinking": "Reasoning"
606
  },
607
  {
 
612
  "Result": 45.96,
613
  "Model URL": "https://huggingface.co/google",
614
  "Params": null,
615
+ "Release": "V4",
616
  "Thinking": "Reasoning"
617
  },
618
  {
 
623
  "Result": 69.01,
624
  "Model URL": "https://huggingface.co/google",
625
  "Params": null,
626
+ "Release": "V4",
627
  "Thinking": "Reasoning"
628
  },
629
  {
 
634
  "Result": 50.51,
635
  "Model URL": "https://huggingface.co/google",
636
  "Params": null,
637
+ "Release": "V4",
638
  "Thinking": "Reasoning"
639
  },
640
  {
 
645
  "Result": 67.11,
646
  "Model URL": "https://huggingface.co/google",
647
  "Params": null,
648
+ "Release": "V4",
649
  "Thinking": "Reasoning"
650
  },
651
  {
 
656
  "Result": 44.66,
657
  "Model URL": "https://huggingface.co/google",
658
  "Params": null,
659
+ "Release": "V4",
660
  "Thinking": "Reasoning"
661
  },
662
  {
 
667
  "Result": 68.68,
668
  "Model URL": "https://huggingface.co/google",
669
  "Params": null,
670
+ "Release": "V4",
671
  "Thinking": "Reasoning"
672
  },
673
  {
 
678
  "Result": 48.79,
679
  "Model URL": "https://huggingface.co/google",
680
  "Params": null,
681
+ "Release": "V4",
682
  "Thinking": "Reasoning"
683
  },
684
  {
 
689
  "Result": -1.0,
690
  "Model URL": "https://huggingface.co/google",
691
  "Params": null,
692
+ "Release": "V4",
693
  "Thinking": "Reasoning"
694
  },
695
  {
 
700
  "Result": 91.03,
701
  "Model URL": "https://huggingface.co/google",
702
  "Params": null,
703
+ "Release": "V4",
704
  "Thinking": "Reasoning"
705
  },
706
  {
 
711
  "Result": 95.29,
712
  "Model URL": "https://huggingface.co/google",
713
  "Params": null,
714
+ "Release": "V4",
715
  "Thinking": "Reasoning"
716
  },
717
  {
 
722
  "Result": 71.79,
723
  "Model URL": "https://huggingface.co/google",
724
  "Params": null,
725
+ "Release": "V4",
726
  "Thinking": "Reasoning"
727
  },
728
  {
 
733
  "Result": 77.65,
734
  "Model URL": "https://huggingface.co/google",
735
  "Params": null,
736
+ "Release": "V4",
737
  "Thinking": "Reasoning"
738
  },
739
  {
 
744
  "Result": 71.41,
745
  "Model URL": "https://huggingface.co/google",
746
  "Params": null,
747
+ "Release": "V4",
748
  "Thinking": "Reasoning"
749
  },
750
  {
 
755
  "Result": 77.65,
756
  "Model URL": "https://huggingface.co/google",
757
  "Params": null,
758
+ "Release": "V4",
759
  "Thinking": "Reasoning"
760
  },
761
  {
 
766
  "Result": 69.91,
767
  "Model URL": "https://huggingface.co/google",
768
  "Params": null,
769
+ "Release": "V4",
770
  "Thinking": "Reasoning"
771
  },
772
  {
 
777
  "Result": 71.27,
778
  "Model URL": "https://huggingface.co/google",
779
  "Params": null,
780
+ "Release": "V4",
781
  "Thinking": "Reasoning"
782
  },
783
  {
 
788
  "Result": 68.04,
789
  "Model URL": "https://huggingface.co/google",
790
  "Params": null,
791
+ "Release": "V4",
792
  "Thinking": "Reasoning"
793
  },
794
  {
 
799
  "Result": 77.35,
800
  "Model URL": "https://huggingface.co/google",
801
  "Params": null,
802
+ "Release": "V4",
803
  "Thinking": "Reasoning"
804
  },
805
  {
 
810
  "Result": 69.52,
811
  "Model URL": "https://huggingface.co/google",
812
  "Params": null,
813
+ "Release": "V4",
814
  "Thinking": "Reasoning"
815
  },
816
  {
 
821
  "Result": 79.45,
822
  "Model URL": "https://huggingface.co/google",
823
  "Params": null,
824
+ "Release": "V4",
825
  "Thinking": "Reasoning"
826
  },
827
  {
 
3032
  "Result": 88.97,
3033
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3034
  "Params": 36.2,
3035
+ "Release": "V4",
3036
  "Thinking": "Reasoning"
3037
  },
3038
  {
 
3043
  "Result": 81.7,
3044
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3045
  "Params": 36.2,
3046
+ "Release": "V4",
3047
  "Thinking": "Reasoning"
3048
  },
3049
  {
 
3054
  "Result": 71.15,
3055
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3056
  "Params": 36.2,
3057
+ "Release": "V4",
3058
  "Thinking": "Reasoning"
3059
  },
3060
  {
 
3065
  "Result": 59.15,
3066
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3067
  "Params": 36.2,
3068
+ "Release": "V4",
3069
  "Thinking": "Reasoning"
3070
  },
3071
  {
 
3076
  "Result": 71.15,
3077
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3078
  "Params": 36.2,
3079
+ "Release": "V4",
3080
  "Thinking": "Reasoning"
3081
  },
3082
  {
 
3087
  "Result": 56.6,
3088
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3089
  "Params": 36.2,
3090
+ "Release": "V4",
3091
  "Thinking": "Reasoning"
3092
  },
3093
  {
 
3098
  "Result": 70.1,
3099
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3100
  "Params": 36.2,
3101
+ "Release": "V4",
3102
  "Thinking": "Reasoning"
3103
  },
3104
  {
 
3109
  "Result": 64.63,
3110
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3111
  "Params": 36.2,
3112
+ "Release": "V4",
3113
  "Thinking": "Reasoning"
3114
  },
3115
  {
 
3120
  "Result": 68.1,
3121
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3122
  "Params": 36.2,
3123
+ "Release": "V4",
3124
  "Thinking": "Reasoning"
3125
  },
3126
  {
 
3131
  "Result": 57.32,
3132
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3133
  "Params": 36.2,
3134
+ "Release": "V4",
3135
  "Thinking": "Reasoning"
3136
  },
3137
  {
 
3142
  "Result": 70.32,
3143
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3144
  "Params": 36.2,
3145
+ "Release": "V4",
3146
  "Thinking": "Reasoning"
3147
  },
3148
  {
 
3153
  "Result": 60.88,
3154
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3155
  "Params": 36.2,
3156
+ "Release": "V4",
3157
  "Thinking": "Reasoning"
3158
  },
3159
  {
 
3164
  "Result": -1.0,
3165
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3166
  "Params": 36.2,
3167
+ "Release": "V4",
3168
  "Thinking": "Reasoning"
3169
  },
3170
  {
 
3175
  "Result": 91.03,
3176
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3177
  "Params": 36.2,
3178
+ "Release": "V4",
3179
  "Thinking": "Reasoning"
3180
  },
3181
  {
 
3186
  "Result": 91.76,
3187
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3188
  "Params": 36.2,
3189
+ "Release": "V4",
3190
  "Thinking": "Reasoning"
3191
  },
3192
  {
 
3197
  "Result": 76.54,
3198
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3199
  "Params": 36.2,
3200
+ "Release": "V4",
3201
  "Thinking": "Reasoning"
3202
  },
3203
  {
 
3208
  "Result": 70.59,
3209
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3210
  "Params": 36.2,
3211
+ "Release": "V4",
3212
  "Thinking": "Reasoning"
3213
  },
3214
  {
 
3219
  "Result": 76.15,
3220
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3221
  "Params": 36.2,
3222
+ "Release": "V4",
3223
  "Thinking": "Reasoning"
3224
  },
3225
  {
 
3230
  "Result": 70.59,
3231
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3232
  "Params": 36.2,
3233
+ "Release": "V4",
3234
  "Thinking": "Reasoning"
3235
  },
3236
  {
 
3241
  "Result": 74.95,
3242
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3243
  "Params": 36.2,
3244
+ "Release": "V4",
3245
  "Thinking": "Reasoning"
3246
  },
3247
  {
 
3252
  "Result": 66.11,
3253
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3254
  "Params": 36.2,
3255
+ "Release": "V4",
3256
  "Thinking": "Reasoning"
3257
  },
3258
  {
 
3263
  "Result": 73.33,
3264
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3265
  "Params": 36.2,
3266
+ "Release": "V4",
3267
  "Thinking": "Reasoning"
3268
  },
3269
  {
 
3274
  "Result": 70.19,
3275
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3276
  "Params": 36.2,
3277
+ "Release": "V4",
3278
  "Thinking": "Reasoning"
3279
  },
3280
  {
 
3285
  "Result": 74.69,
3286
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3287
  "Params": 36.2,
3288
+ "Release": "V4",
3289
  "Thinking": "Reasoning"
3290
  },
3291
  {
 
3296
  "Result": 72.45,
3297
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3298
  "Params": 36.2,
3299
+ "Release": "V4",
3300
  "Thinking": "Reasoning"
3301
  },
3302
  {
 
4957
  "Result": 79.74,
4958
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4959
  "Params": 14,
4960
+ "Release": "V4",
4961
  "Thinking": "Reasoning"
4962
  },
4963
  {
 
4968
  "Result": 76.6,
4969
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4970
  "Params": 14,
4971
+ "Release": "V4",
4972
  "Thinking": "Reasoning"
4973
  },
4974
  {
 
4979
  "Result": 55.9,
4980
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4981
  "Params": 14,
4982
+ "Release": "V4",
4983
  "Thinking": "Reasoning"
4984
  },
4985
  {
 
4990
  "Result": 39.57,
4991
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4992
  "Params": 14,
4993
+ "Release": "V4",
4994
  "Thinking": "Reasoning"
4995
  },
4996
  {
 
5001
  "Result": 55.64,
5002
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5003
  "Params": 14,
5004
+ "Release": "V4",
5005
  "Thinking": "Reasoning"
5006
  },
5007
  {
 
5012
  "Result": 34.47,
5013
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5014
  "Params": 14,
5015
+ "Release": "V4",
5016
  "Thinking": "Reasoning"
5017
  },
5018
  {
 
5023
  "Result": 55.34,
5024
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5025
  "Params": 14,
5026
+ "Release": "V4",
5027
  "Thinking": "Reasoning"
5028
  },
5029
  {
 
5034
  "Result": 38.35,
5035
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5036
  "Params": 14,
5037
+ "Release": "V4",
5038
  "Thinking": "Reasoning"
5039
  },
5040
  {
 
5045
  "Result": 53.73,
5046
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5047
  "Params": 14,
5048
+ "Release": "V4",
5049
  "Thinking": "Reasoning"
5050
  },
5051
  {
 
5056
  "Result": 32.73,
5057
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5058
  "Params": 14,
5059
+ "Release": "V4",
5060
  "Thinking": "Reasoning"
5061
  },
5062
  {
 
5067
  "Result": 55.17,
5068
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5069
  "Params": 14,
5070
+ "Release": "V4",
5071
  "Thinking": "Reasoning"
5072
  },
5073
  {
 
5078
  "Result": 36.13,
5079
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5080
  "Params": 14,
5081
+ "Release": "V4",
5082
  "Thinking": "Reasoning"
5083
  },
5084
  {
 
5089
  "Result": -1.0,
5090
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5091
  "Params": 14,
5092
+ "Release": "V4",
5093
  "Thinking": "Reasoning"
5094
  },
5095
  {
 
5100
  "Result": 51.15,
5101
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5102
  "Params": 14,
5103
+ "Release": "V4",
5104
  "Thinking": "Reasoning"
5105
  },
5106
  {
 
5111
  "Result": 74.12,
5112
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5113
  "Params": 14,
5114
+ "Release": "V4",
5115
  "Thinking": "Reasoning"
5116
  },
5117
  {
 
5122
  "Result": 36.54,
5123
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5124
  "Params": 14,
5125
+ "Release": "V4",
5126
  "Thinking": "Reasoning"
5127
  },
5128
  {
 
5133
  "Result": 56.47,
5134
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5135
  "Params": 14,
5136
+ "Release": "V4",
5137
  "Thinking": "Reasoning"
5138
  },
5139
  {
 
5144
  "Result": 36.28,
5145
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5146
  "Params": 14,
5147
+ "Release": "V4",
5148
  "Thinking": "Reasoning"
5149
  },
5150
  {
 
5155
  "Result": 56.47,
5156
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5157
  "Params": 14,
5158
+ "Release": "V4",
5159
  "Thinking": "Reasoning"
5160
  },
5161
  {
 
5166
  "Result": 35.26,
5167
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5168
  "Params": 14,
5169
+ "Release": "V4",
5170
  "Thinking": "Reasoning"
5171
  },
5172
  {
 
5177
  "Result": 52.97,
5178
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5179
  "Params": 14,
5180
+ "Release": "V4",
5181
  "Thinking": "Reasoning"
5182
  },
5183
  {
 
5188
  "Result": 35.52,
5189
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5190
  "Params": 14,
5191
+ "Release": "V4",
5192
  "Thinking": "Reasoning"
5193
  },
5194
  {
 
5199
  "Result": 56.22,
5200
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5201
  "Params": 14,
5202
+ "Release": "V4",
5203
  "Thinking": "Reasoning"
5204
  },
5205
  {
 
5210
  "Result": 34.94,
5211
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5212
  "Params": 14,
5213
+ "Release": "V4",
5214
  "Thinking": "Reasoning"
5215
  },
5216
  {
 
5221
  "Result": 57.44,
5222
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5223
  "Params": 14,
5224
+ "Release": "V4",
5225
  "Thinking": "Reasoning"
5226
  },
5227
  {
 
5232
  "Result": 72.18,
5233
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5234
  "Params": 14,
5235
+ "Release": "V4",
5236
  "Thinking": "Dense"
5237
  },
5238
  {
 
5243
  "Result": 80.85,
5244
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5245
  "Params": 14,
5246
+ "Release": "V4",
5247
  "Thinking": "Dense"
5248
  },
5249
  {
 
5254
  "Result": 42.31,
5255
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5256
  "Params": 14,
5257
+ "Release": "V4",
5258
  "Thinking": "Dense"
5259
  },
5260
  {
 
5265
  "Result": 48.09,
5266
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5267
  "Params": 14,
5268
+ "Release": "V4",
5269
  "Thinking": "Dense"
5270
  },
5271
  {
 
5276
  "Result": 42.31,
5277
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5278
  "Params": 14,
5279
+ "Release": "V4",
5280
  "Thinking": "Dense"
5281
  },
5282
  {
 
5287
  "Result": 45.53,
5288
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5289
  "Params": 14,
5290
+ "Release": "V4",
5291
  "Thinking": "Dense"
5292
  },
5293
  {
 
5298
  "Result": 41.73,
5299
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5300
  "Params": 14,
5301
+ "Release": "V4",
5302
  "Thinking": "Dense"
5303
  },
5304
  {
 
5309
  "Result": 50.59,
5310
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5311
  "Params": 14,
5312
+ "Release": "V4",
5313
  "Thinking": "Dense"
5314
  },
5315
  {
 
5320
  "Result": 40.33,
5321
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5322
  "Params": 14,
5323
+ "Release": "V4",
5324
  "Thinking": "Dense"
5325
  },
5326
  {
 
5331
  "Result": 45.81,
5332
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5333
  "Params": 14,
5334
+ "Release": "V4",
5335
  "Thinking": "Dense"
5336
  },
5337
  {
 
5342
  "Result": 41.38,
5343
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5344
  "Params": 14,
5345
+ "Release": "V4",
5346
  "Thinking": "Dense"
5347
  },
5348
  {
 
5353
  "Result": 47.91,
5354
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5355
  "Params": 14,
5356
+ "Release": "V4",
5357
  "Thinking": "Dense"
5358
  },
5359
  {
 
5364
  "Result": 28.36,
5365
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5366
  "Params": 14,
5367
+ "Release": "V4",
5368
  "Thinking": "Dense"
5369
  },
5370
  {
 
5375
  "Result": 52.69,
5376
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5377
  "Params": 14,
5378
+ "Release": "V4",
5379
  "Thinking": "Dense"
5380
  },
5381
  {
 
5386
  "Result": 83.53,
5387
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5388
  "Params": 14,
5389
+ "Release": "V4",
5390
  "Thinking": "Dense"
5391
  },
5392
  {
 
5397
  "Result": 27.44,
5398
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5399
  "Params": 14,
5400
+ "Release": "V4",
5401
  "Thinking": "Dense"
5402
  },
5403
  {
 
5408
  "Result": 45.88,
5409
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5410
  "Params": 14,
5411
+ "Release": "V4",
5412
  "Thinking": "Dense"
5413
  },
5414
  {
 
5419
  "Result": 27.44,
5420
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5421
  "Params": 14,
5422
+ "Release": "V4",
5423
  "Thinking": "Dense"
5424
  },
5425
  {
 
5430
  "Result": 45.88,
5431
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5432
  "Params": 14,
5433
+ "Release": "V4",
5434
  "Thinking": "Dense"
5435
  },
5436
  {
 
5441
  "Result": 26.54,
5442
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5443
  "Params": 14,
5444
+ "Release": "V4",
5445
  "Thinking": "Dense"
5446
  },
5447
  {
 
5452
  "Result": 43.55,
5453
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5454
  "Params": 14,
5455
+ "Release": "V4",
5456
  "Thinking": "Dense"
5457
  },
5458
  {
 
5463
  "Result": 26.68,
5464
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5465
  "Params": 14,
5466
+ "Release": "V4",
5467
  "Thinking": "Dense"
5468
  },
5469
  {
 
5474
  "Result": 45.98,
5475
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5476
  "Params": 14,
5477
+ "Release": "V4",
5478
  "Thinking": "Dense"
5479
  },
5480
  {
 
5485
  "Result": 25.11,
5486
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5487
  "Params": 14,
5488
+ "Release": "V4",
5489
  "Thinking": "Dense"
5490
  },
5491
  {
 
5496
  "Result": 44.96,
5497
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5498
  "Params": 14,
5499
+ "Release": "V4",
5500
  "Thinking": "Dense"
5501
  },
5502
  {
 
5507
  "Result": 70.77,
5508
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5509
  "Params": 8.2,
5510
+ "Release": "V4",
5511
  "Thinking": "Reasoning"
5512
  },
5513
  {
 
5518
  "Result": 62.13,
5519
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5520
  "Params": 8.2,
5521
+ "Release": "V4",
5522
  "Thinking": "Reasoning"
5523
  },
5524
  {
 
5529
  "Result": 50.0,
5530
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5531
  "Params": 8.2,
5532
+ "Release": "V4",
5533
  "Thinking": "Reasoning"
5534
  },
5535
  {
 
5540
  "Result": 34.47,
5541
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5542
  "Params": 8.2,
5543
+ "Release": "V4",
5544
  "Thinking": "Reasoning"
5545
  },
5546
  {
 
5551
  "Result": 50.0,
5552
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5553
  "Params": 8.2,
5554
+ "Release": "V4",
5555
  "Thinking": "Reasoning"
5556
  },
5557
  {
 
5562
  "Result": 32.77,
5563
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5564
  "Params": 8.2,
5565
+ "Release": "V4",
5566
  "Thinking": "Reasoning"
5567
  },
5568
  {
 
5573
  "Result": 49.33,
5574
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5575
  "Params": 8.2,
5576
+ "Release": "V4",
5577
  "Thinking": "Reasoning"
5578
  },
5579
  {
 
5584
  "Result": 35.1,
5585
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5586
  "Params": 8.2,
5587
+ "Release": "V4",
5588
  "Thinking": "Reasoning"
5589
  },
5590
  {
 
5595
  "Result": 47.92,
5596
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5597
  "Params": 8.2,
5598
+ "Release": "V4",
5599
  "Thinking": "Reasoning"
5600
  },
5601
  {
 
5606
  "Result": 32.52,
5607
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5608
  "Params": 8.2,
5609
+ "Release": "V4",
5610
  "Thinking": "Reasoning"
5611
  },
5612
  {
 
5617
  "Result": 48.91,
5618
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5619
  "Params": 8.2,
5620
+ "Release": "V4",
5621
  "Thinking": "Reasoning"
5622
  },
5623
  {
 
5628
  "Result": 31.88,
5629
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5630
  "Params": 8.2,
5631
+ "Release": "V4",
5632
  "Thinking": "Reasoning"
5633
  },
5634
  {
 
5639
  "Result": -1.0,
5640
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5641
  "Params": 8.2,
5642
+ "Release": "V4",
5643
  "Thinking": "Reasoning"
5644
  },
5645
  {
 
5650
  "Result": 69.36,
5651
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5652
  "Params": 8.2,
5653
+ "Release": "V4",
5654
  "Thinking": "Reasoning"
5655
  },
5656
  {
 
5661
  "Result": 81.18,
5662
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5663
  "Params": 8.2,
5664
+ "Release": "V4",
5665
  "Thinking": "Reasoning"
5666
  },
5667
  {
 
5672
  "Result": 51.54,
5673
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5674
  "Params": 8.2,
5675
+ "Release": "V4",
5676
  "Thinking": "Reasoning"
5677
  },
5678
  {
 
5683
  "Result": 40.0,
5684
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5685
  "Params": 8.2,
5686
+ "Release": "V4",
5687
  "Thinking": "Reasoning"
5688
  },
5689
  {
 
5694
  "Result": 50.64,
5695
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5696
  "Params": 8.2,
5697
+ "Release": "V4",
5698
  "Thinking": "Reasoning"
5699
  },
5700
  {
 
5705
  "Result": 38.82,
5706
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5707
  "Params": 8.2,
5708
+ "Release": "V4",
5709
  "Thinking": "Reasoning"
5710
  },
5711
  {
 
5716
  "Result": 49.91,
5717
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5718
  "Params": 8.2,
5719
+ "Release": "V4",
5720
  "Thinking": "Reasoning"
5721
  },
5722
  {
 
5727
  "Result": 37.47,
5728
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5729
  "Params": 8.2,
5730
+ "Release": "V4",
5731
  "Thinking": "Reasoning"
5732
  },
5733
  {
 
5738
  "Result": 48.4,
5739
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5740
  "Params": 8.2,
5741
+ "Release": "V4",
5742
  "Thinking": "Reasoning"
5743
  },
5744
  {
 
5749
  "Result": 38.76,
5750
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5751
  "Params": 8.2,
5752
+ "Release": "V4",
5753
  "Thinking": "Reasoning"
5754
  },
5755
  {
 
5760
  "Result": 49.41,
5761
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5762
  "Params": 8.2,
5763
+ "Release": "V4",
5764
  "Thinking": "Reasoning"
5765
  },
5766
  {
 
5771
  "Result": 38.73,
5772
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5773
  "Params": 8.2,
5774
+ "Release": "V4",
5775
  "Thinking": "Reasoning"
5776
  },
5777
  {
results/results_icarus_november_2025.csv CHANGED
@@ -1,5 +1,7 @@
1
  ,Syntax (STX),Syntax (STX),Functionality (FNC),Functionality (FNC),Synthesis (SYN),Synthesis (SYN),Power,Power,Performance,Performance,Area,Area,EM,Syntax (STX),Syntax (STX),Functionality (FNC),Functionality (FNC),Synthesis (SYN),Synthesis (SYN),Power,Power,Performance,Performance,Area,Area
2
  ,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,RTL-Repo,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen
 
 
3
  Gemini 2.5 Flash (Medium),92.049999999999997,79.569999999999993,70.379999999999995,62.130000000000003,70.260000000000005,45.960000000000001,69.010000000000005,50.509999999999998,67.109999999999999,44.659999999999997,68.680000000000007,48.789999999999999,-1,91.030000000000001,95.290000000000006,71.790000000000006,77.650000000000006,71.409999999999997,77.650000000000006,69.909999999999997,71.269999999999996,68.040000000000006,77.349999999999994,69.519999999999996,79.450000000000003
4
  DeepSeek R1-0528,96.150000000000006,86.120000000000005,81.540000000000006,64.900000000000006,81.280000000000001,64.489999999999995,79.150000000000006,73.329999999999998,78.090000000000003,65.640000000000001,79.280000000000001,71.150000000000006,-1,95.379999999999995,91.760000000000005,81.540000000000006,74.120000000000005,81.409999999999997,74.120000000000005,80.269999999999996,63.859999999999999,78.689999999999998,74.049999999999997,79.989999999999995,77.010000000000005
5
  DeepSeek R1,97.180000000000007,89.799999999999997,79.739999999999995,65.709999999999994,79.620000000000005,63.270000000000003,78.329999999999998,71.340000000000003,76.489999999999995,64.060000000000002,78.189999999999998,70.079999999999998,-1,97.439999999999998,96.469999999999999,79.489999999999995,74.120000000000005,79.489999999999995,74.120000000000005,78.269999999999996,64.340000000000003,76.430000000000007,74.290000000000006,77.959999999999994,77.120000000000005
@@ -65,5 +67,4 @@ CodeV-DS-6.7B,33.590000000000003,67.349999999999994,15,38.780000000000001,15,37.
65
  ,,,,,,,,,,,,,,,,,,,,,,,,,
66
  ,,,,,,,,,,,,,,,,,,,,,,,,,
67
  ,,,,,,,,,,,,,,,,,,,,,,,,,
68
- ,,,,,,,,,,,,,,,,,,,,,,,,,
69
 
 
1
  ,Syntax (STX),Syntax (STX),Functionality (FNC),Functionality (FNC),Synthesis (SYN),Synthesis (SYN),Power,Power,Performance,Performance,Area,Area,EM,Syntax (STX),Syntax (STX),Functionality (FNC),Functionality (FNC),Synthesis (SYN),Synthesis (SYN),Power,Power,Performance,Performance,Area,Area
2
  ,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,RTL-Repo,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen
3
+ DeepSeek V3.1 Terminus,96.280000000000001,90.209999999999994,76.790000000000006,60.43,76.670000000000002,58.299999999999997,76.049999999999997,63.840000000000003,73.760000000000005,58.770000000000003,74.939999999999998,61.829999999999998,-1,95.379999999999995,92.939999999999998,78.969999999999999,75.290000000000006,78.849999999999994,75.290000000000006,77.340000000000003,69.870000000000005,76.109999999999999,75.140000000000001,77.030000000000001,77.590000000000003
4
+ Kimi K2 Instruct 0905,96.150000000000006,87.230000000000004,72.819999999999993,62.130000000000003,72.689999999999998,60.43,71.280000000000001,63.479999999999997,69.980000000000004,58.579999999999998,71.200000000000003,63.32,33.649999999999999,98.209999999999994,95.290000000000006,74.739999999999995,69.409999999999997,74.489999999999995,69.409999999999997,73.129999999999995,57.469999999999999,71.430000000000007,69.900000000000006,72.760000000000005,69.510000000000005
5
  Gemini 2.5 Flash (Medium),92.049999999999997,79.569999999999993,70.379999999999995,62.130000000000003,70.260000000000005,45.960000000000001,69.010000000000005,50.509999999999998,67.109999999999999,44.659999999999997,68.680000000000007,48.789999999999999,-1,91.030000000000001,95.290000000000006,71.790000000000006,77.650000000000006,71.409999999999997,77.650000000000006,69.909999999999997,71.269999999999996,68.040000000000006,77.349999999999994,69.519999999999996,79.450000000000003
6
  DeepSeek R1-0528,96.150000000000006,86.120000000000005,81.540000000000006,64.900000000000006,81.280000000000001,64.489999999999995,79.150000000000006,73.329999999999998,78.090000000000003,65.640000000000001,79.280000000000001,71.150000000000006,-1,95.379999999999995,91.760000000000005,81.540000000000006,74.120000000000005,81.409999999999997,74.120000000000005,80.269999999999996,63.859999999999999,78.689999999999998,74.049999999999997,79.989999999999995,77.010000000000005
7
  DeepSeek R1,97.180000000000007,89.799999999999997,79.739999999999995,65.709999999999994,79.620000000000005,63.270000000000003,78.329999999999998,71.340000000000003,76.489999999999995,64.060000000000002,78.189999999999998,70.079999999999998,-1,97.439999999999998,96.469999999999999,79.489999999999995,74.120000000000005,79.489999999999995,74.120000000000005,78.269999999999996,64.340000000000003,76.430000000000007,74.290000000000006,77.959999999999994,77.120000000000005
 
67
  ,,,,,,,,,,,,,,,,,,,,,,,,,
68
  ,,,,,,,,,,,,,,,,,,,,,,,,,
69
  ,,,,,,,,,,,,,,,,,,,,,,,,,
 
70
 
results/results_verilator.json CHANGED
@@ -1,4 +1,554 @@
1
  [
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  {
3
  "Model": "Gemini 2.5 Flash (Medium)",
4
  "Model Type": "General",
@@ -7,7 +557,7 @@
7
  "Result": 91.54,
8
  "Model URL": "https://huggingface.co/google",
9
  "Params": null,
10
- "Release": "V3",
11
  "Thinking": "Reasoning"
12
  },
13
  {
@@ -18,7 +568,7 @@
18
  "Result": 81.7,
19
  "Model URL": "https://huggingface.co/google",
20
  "Params": null,
21
- "Release": "V3",
22
  "Thinking": "Reasoning"
23
  },
24
  {
@@ -29,7 +579,7 @@
29
  "Result": 69.74,
30
  "Model URL": "https://huggingface.co/google",
31
  "Params": null,
32
- "Release": "V3",
33
  "Thinking": "Reasoning"
34
  },
35
  {
@@ -40,7 +590,7 @@
40
  "Result": 64.68,
41
  "Model URL": "https://huggingface.co/google",
42
  "Params": null,
43
- "Release": "V3",
44
  "Thinking": "Reasoning"
45
  },
46
  {
@@ -51,7 +601,7 @@
51
  "Result": 69.23,
52
  "Model URL": "https://huggingface.co/google",
53
  "Params": null,
54
- "Release": "V3",
55
  "Thinking": "Reasoning"
56
  },
57
  {
@@ -62,7 +612,7 @@
62
  "Result": 46.38,
63
  "Model URL": "https://huggingface.co/google",
64
  "Params": null,
65
- "Release": "V3",
66
  "Thinking": "Reasoning"
67
  },
68
  {
@@ -73,7 +623,7 @@
73
  "Result": 68.4,
74
  "Model URL": "https://huggingface.co/google",
75
  "Params": null,
76
- "Release": "V3",
77
  "Thinking": "Reasoning"
78
  },
79
  {
@@ -84,7 +634,7 @@
84
  "Result": 51.11,
85
  "Model URL": "https://huggingface.co/google",
86
  "Params": null,
87
- "Release": "V3",
88
  "Thinking": "Reasoning"
89
  },
90
  {
@@ -95,7 +645,7 @@
95
  "Result": 66.57,
96
  "Model URL": "https://huggingface.co/google",
97
  "Params": null,
98
- "Release": "V3",
99
  "Thinking": "Reasoning"
100
  },
101
  {
@@ -106,7 +656,7 @@
106
  "Result": 45.1,
107
  "Model URL": "https://huggingface.co/google",
108
  "Params": null,
109
- "Release": "V3",
110
  "Thinking": "Reasoning"
111
  },
112
  {
@@ -117,7 +667,7 @@
117
  "Result": 68.27,
118
  "Model URL": "https://huggingface.co/google",
119
  "Params": null,
120
- "Release": "V3",
121
  "Thinking": "Reasoning"
122
  },
123
  {
@@ -128,7 +678,7 @@
128
  "Result": 49.25,
129
  "Model URL": "https://huggingface.co/google",
130
  "Params": null,
131
- "Release": "V3",
132
  "Thinking": "Reasoning"
133
  },
134
  {
@@ -139,7 +689,7 @@
139
  "Result": -1.0,
140
  "Model URL": "https://huggingface.co/google",
141
  "Params": null,
142
- "Release": "V3",
143
  "Thinking": "Reasoning"
144
  },
145
  {
@@ -150,7 +700,7 @@
150
  "Result": 90.64,
151
  "Model URL": "https://huggingface.co/google",
152
  "Params": null,
153
- "Release": "V3",
154
  "Thinking": "Reasoning"
155
  },
156
  {
@@ -161,7 +711,7 @@
161
  "Result": 96.47,
162
  "Model URL": "https://huggingface.co/google",
163
  "Params": null,
164
- "Release": "V3",
165
  "Thinking": "Reasoning"
166
  },
167
  {
@@ -172,7 +722,7 @@
172
  "Result": 71.67,
173
  "Model URL": "https://huggingface.co/google",
174
  "Params": null,
175
- "Release": "V3",
176
  "Thinking": "Reasoning"
177
  },
178
  {
@@ -183,7 +733,7 @@
183
  "Result": 77.65,
184
  "Model URL": "https://huggingface.co/google",
185
  "Params": null,
186
- "Release": "V3",
187
  "Thinking": "Reasoning"
188
  },
189
  {
@@ -194,7 +744,7 @@
194
  "Result": 71.28,
195
  "Model URL": "https://huggingface.co/google",
196
  "Params": null,
197
- "Release": "V3",
198
  "Thinking": "Reasoning"
199
  },
200
  {
@@ -205,7 +755,7 @@
205
  "Result": 77.65,
206
  "Model URL": "https://huggingface.co/google",
207
  "Params": null,
208
- "Release": "V3",
209
  "Thinking": "Reasoning"
210
  },
211
  {
@@ -216,7 +766,7 @@
216
  "Result": 70.28,
217
  "Model URL": "https://huggingface.co/google",
218
  "Params": null,
219
- "Release": "V3",
220
  "Thinking": "Reasoning"
221
  },
222
  {
@@ -227,7 +777,7 @@
227
  "Result": 71.27,
228
  "Model URL": "https://huggingface.co/google",
229
  "Params": null,
230
- "Release": "V3",
231
  "Thinking": "Reasoning"
232
  },
233
  {
@@ -238,7 +788,7 @@
238
  "Result": 68.06,
239
  "Model URL": "https://huggingface.co/google",
240
  "Params": null,
241
- "Release": "V3",
242
  "Thinking": "Reasoning"
243
  },
244
  {
@@ -249,7 +799,7 @@
249
  "Result": 77.35,
250
  "Model URL": "https://huggingface.co/google",
251
  "Params": null,
252
- "Release": "V3",
253
  "Thinking": "Reasoning"
254
  },
255
  {
@@ -260,7 +810,7 @@
260
  "Result": 70.32,
261
  "Model URL": "https://huggingface.co/google",
262
  "Params": null,
263
- "Release": "V3",
264
  "Thinking": "Reasoning"
265
  },
266
  {
@@ -271,7 +821,7 @@
271
  "Result": 79.45,
272
  "Model URL": "https://huggingface.co/google",
273
  "Params": null,
274
- "Release": "V3",
275
  "Thinking": "Reasoning"
276
  },
277
  {
@@ -2482,7 +3032,7 @@
2482
  "Result": 89.36,
2483
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2484
  "Params": 36.2,
2485
- "Release": "V3",
2486
  "Thinking": "Reasoning"
2487
  },
2488
  {
@@ -2493,7 +3043,7 @@
2493
  "Result": 83.4,
2494
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2495
  "Params": 36.2,
2496
- "Release": "V3",
2497
  "Thinking": "Reasoning"
2498
  },
2499
  {
@@ -2504,7 +3054,7 @@
2504
  "Result": 71.15,
2505
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2506
  "Params": 36.2,
2507
- "Release": "V3",
2508
  "Thinking": "Reasoning"
2509
  },
2510
  {
@@ -2515,7 +3065,7 @@
2515
  "Result": 61.7,
2516
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2517
  "Params": 36.2,
2518
- "Release": "V3",
2519
  "Thinking": "Reasoning"
2520
  },
2521
  {
@@ -2526,7 +3076,7 @@
2526
  "Result": 70.9,
2527
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2528
  "Params": 36.2,
2529
- "Release": "V3",
2530
  "Thinking": "Reasoning"
2531
  },
2532
  {
@@ -2537,7 +3087,7 @@
2537
  "Result": 57.02,
2538
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2539
  "Params": 36.2,
2540
- "Release": "V3",
2541
  "Thinking": "Reasoning"
2542
  },
2543
  {
@@ -2548,7 +3098,7 @@
2548
  "Result": 70.53,
2549
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2550
  "Params": 36.2,
2551
- "Release": "V3",
2552
  "Thinking": "Reasoning"
2553
  },
2554
  {
@@ -2559,7 +3109,7 @@
2559
  "Result": 65.33,
2560
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2561
  "Params": 36.2,
2562
- "Release": "V3",
2563
  "Thinking": "Reasoning"
2564
  },
2565
  {
@@ -2570,7 +3120,7 @@
2570
  "Result": 68.22,
2571
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2572
  "Params": 36.2,
2573
- "Release": "V3",
2574
  "Thinking": "Reasoning"
2575
  },
2576
  {
@@ -2581,7 +3131,7 @@
2581
  "Result": 57.32,
2582
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2583
  "Params": 36.2,
2584
- "Release": "V3",
2585
  "Thinking": "Reasoning"
2586
  },
2587
  {
@@ -2592,7 +3142,7 @@
2592
  "Result": 70.33,
2593
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2594
  "Params": 36.2,
2595
- "Release": "V3",
2596
  "Thinking": "Reasoning"
2597
  },
2598
  {
@@ -2603,7 +3153,7 @@
2603
  "Result": 61.39,
2604
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2605
  "Params": 36.2,
2606
- "Release": "V3",
2607
  "Thinking": "Reasoning"
2608
  },
2609
  {
@@ -2614,7 +3164,7 @@
2614
  "Result": -1.0,
2615
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2616
  "Params": 36.2,
2617
- "Release": "V3",
2618
  "Thinking": "Reasoning"
2619
  },
2620
  {
@@ -2625,7 +3175,7 @@
2625
  "Result": 91.03,
2626
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2627
  "Params": 36.2,
2628
- "Release": "V3",
2629
  "Thinking": "Reasoning"
2630
  },
2631
  {
@@ -2636,7 +3186,7 @@
2636
  "Result": 91.76,
2637
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2638
  "Params": 36.2,
2639
- "Release": "V3",
2640
  "Thinking": "Reasoning"
2641
  },
2642
  {
@@ -2647,7 +3197,7 @@
2647
  "Result": 73.85,
2648
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2649
  "Params": 36.2,
2650
- "Release": "V3",
2651
  "Thinking": "Reasoning"
2652
  },
2653
  {
@@ -2658,7 +3208,7 @@
2658
  "Result": 70.59,
2659
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2660
  "Params": 36.2,
2661
- "Release": "V3",
2662
  "Thinking": "Reasoning"
2663
  },
2664
  {
@@ -2669,7 +3219,7 @@
2669
  "Result": 73.46,
2670
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2671
  "Params": 36.2,
2672
- "Release": "V3",
2673
  "Thinking": "Reasoning"
2674
  },
2675
  {
@@ -2680,7 +3230,7 @@
2680
  "Result": 70.59,
2681
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2682
  "Params": 36.2,
2683
- "Release": "V3",
2684
  "Thinking": "Reasoning"
2685
  },
2686
  {
@@ -2691,7 +3241,7 @@
2691
  "Result": 73.04,
2692
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2693
  "Params": 36.2,
2694
- "Release": "V3",
2695
  "Thinking": "Reasoning"
2696
  },
2697
  {
@@ -2702,7 +3252,7 @@
2702
  "Result": 66.11,
2703
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2704
  "Params": 36.2,
2705
- "Release": "V3",
2706
  "Thinking": "Reasoning"
2707
  },
2708
  {
@@ -2713,7 +3263,7 @@
2713
  "Result": 70.89,
2714
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2715
  "Params": 36.2,
2716
- "Release": "V3",
2717
  "Thinking": "Reasoning"
2718
  },
2719
  {
@@ -2724,7 +3274,7 @@
2724
  "Result": 70.19,
2725
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2726
  "Params": 36.2,
2727
- "Release": "V3",
2728
  "Thinking": "Reasoning"
2729
  },
2730
  {
@@ -2735,7 +3285,7 @@
2735
  "Result": 72.6,
2736
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2737
  "Params": 36.2,
2738
- "Release": "V3",
2739
  "Thinking": "Reasoning"
2740
  },
2741
  {
@@ -2746,7 +3296,7 @@
2746
  "Result": 72.45,
2747
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
2748
  "Params": 36.2,
2749
- "Release": "V3",
2750
  "Thinking": "Reasoning"
2751
  },
2752
  {
@@ -4407,7 +4957,7 @@
4407
  "Result": 80.64,
4408
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4409
  "Params": 14,
4410
- "Release": "V3",
4411
  "Thinking": "Reasoning"
4412
  },
4413
  {
@@ -4418,7 +4968,7 @@
4418
  "Result": 79.57,
4419
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4420
  "Params": 14,
4421
- "Release": "V3",
4422
  "Thinking": "Reasoning"
4423
  },
4424
  {
@@ -4429,7 +4979,7 @@
4429
  "Result": 56.15,
4430
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4431
  "Params": 14,
4432
- "Release": "V3",
4433
  "Thinking": "Reasoning"
4434
  },
4435
  {
@@ -4440,7 +4990,7 @@
4440
  "Result": 41.28,
4441
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4442
  "Params": 14,
4443
- "Release": "V3",
4444
  "Thinking": "Reasoning"
4445
  },
4446
  {
@@ -4451,7 +5001,7 @@
4451
  "Result": 55.77,
4452
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4453
  "Params": 14,
4454
- "Release": "V3",
4455
  "Thinking": "Reasoning"
4456
  },
4457
  {
@@ -4462,7 +5012,7 @@
4462
  "Result": 34.47,
4463
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4464
  "Params": 14,
4465
- "Release": "V3",
4466
  "Thinking": "Reasoning"
4467
  },
4468
  {
@@ -4473,7 +5023,7 @@
4473
  "Result": 55.85,
4474
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4475
  "Params": 14,
4476
- "Release": "V3",
4477
  "Thinking": "Reasoning"
4478
  },
4479
  {
@@ -4484,7 +5034,7 @@
4484
  "Result": 38.35,
4485
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4486
  "Params": 14,
4487
- "Release": "V3",
4488
  "Thinking": "Reasoning"
4489
  },
4490
  {
@@ -4495,7 +5045,7 @@
4495
  "Result": 53.71,
4496
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4497
  "Params": 14,
4498
- "Release": "V3",
4499
  "Thinking": "Reasoning"
4500
  },
4501
  {
@@ -4506,7 +5056,7 @@
4506
  "Result": 32.73,
4507
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4508
  "Params": 14,
4509
- "Release": "V3",
4510
  "Thinking": "Reasoning"
4511
  },
4512
  {
@@ -4517,7 +5067,7 @@
4517
  "Result": 55.65,
4518
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4519
  "Params": 14,
4520
- "Release": "V3",
4521
  "Thinking": "Reasoning"
4522
  },
4523
  {
@@ -4528,7 +5078,7 @@
4528
  "Result": 36.13,
4529
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4530
  "Params": 14,
4531
- "Release": "V3",
4532
  "Thinking": "Reasoning"
4533
  },
4534
  {
@@ -4539,7 +5089,7 @@
4539
  "Result": -1.0,
4540
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4541
  "Params": 14,
4542
- "Release": "V3",
4543
  "Thinking": "Reasoning"
4544
  },
4545
  {
@@ -4550,7 +5100,7 @@
4550
  "Result": 52.18,
4551
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4552
  "Params": 14,
4553
- "Release": "V3",
4554
  "Thinking": "Reasoning"
4555
  },
4556
  {
@@ -4561,7 +5111,7 @@
4561
  "Result": 84.71,
4562
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4563
  "Params": 14,
4564
- "Release": "V3",
4565
  "Thinking": "Reasoning"
4566
  },
4567
  {
@@ -4572,7 +5122,7 @@
4572
  "Result": 36.92,
4573
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4574
  "Params": 14,
4575
- "Release": "V3",
4576
  "Thinking": "Reasoning"
4577
  },
4578
  {
@@ -4583,7 +5133,7 @@
4583
  "Result": 60.0,
4584
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4585
  "Params": 14,
4586
- "Release": "V3",
4587
  "Thinking": "Reasoning"
4588
  },
4589
  {
@@ -4594,7 +5144,7 @@
4594
  "Result": 36.67,
4595
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4596
  "Params": 14,
4597
- "Release": "V3",
4598
  "Thinking": "Reasoning"
4599
  },
4600
  {
@@ -4605,7 +5155,7 @@
4605
  "Result": 58.82,
4606
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4607
  "Params": 14,
4608
- "Release": "V3",
4609
  "Thinking": "Reasoning"
4610
  },
4611
  {
@@ -4616,7 +5166,7 @@
4616
  "Result": 36.21,
4617
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4618
  "Params": 14,
4619
- "Release": "V3",
4620
  "Thinking": "Reasoning"
4621
  },
4622
  {
@@ -4627,7 +5177,7 @@
4627
  "Result": 52.97,
4628
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4629
  "Params": 14,
4630
- "Release": "V3",
4631
  "Thinking": "Reasoning"
4632
  },
4633
  {
@@ -4638,7 +5188,7 @@
4638
  "Result": 36.02,
4639
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4640
  "Params": 14,
4641
- "Release": "V3",
4642
  "Thinking": "Reasoning"
4643
  },
4644
  {
@@ -4649,7 +5199,7 @@
4649
  "Result": 58.57,
4650
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4651
  "Params": 14,
4652
- "Release": "V3",
4653
  "Thinking": "Reasoning"
4654
  },
4655
  {
@@ -4660,7 +5210,7 @@
4660
  "Result": 36.12,
4661
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4662
  "Params": 14,
4663
- "Release": "V3",
4664
  "Thinking": "Reasoning"
4665
  },
4666
  {
@@ -4671,7 +5221,7 @@
4671
  "Result": 59.8,
4672
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4673
  "Params": 14,
4674
- "Release": "V3",
4675
  "Thinking": "Reasoning"
4676
  },
4677
  {
@@ -4682,7 +5232,7 @@
4682
  "Result": 72.18,
4683
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4684
  "Params": 14,
4685
- "Release": "V3",
4686
  "Thinking": "Dense"
4687
  },
4688
  {
@@ -4693,7 +5243,7 @@
4693
  "Result": 85.11,
4694
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4695
  "Params": 14,
4696
- "Release": "V3",
4697
  "Thinking": "Dense"
4698
  },
4699
  {
@@ -4704,7 +5254,7 @@
4704
  "Result": 43.85,
4705
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4706
  "Params": 14,
4707
- "Release": "V3",
4708
  "Thinking": "Dense"
4709
  },
4710
  {
@@ -4715,7 +5265,7 @@
4715
  "Result": 50.64,
4716
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4717
  "Params": 14,
4718
- "Release": "V3",
4719
  "Thinking": "Dense"
4720
  },
4721
  {
@@ -4726,7 +5276,7 @@
4726
  "Result": 43.85,
4727
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4728
  "Params": 14,
4729
- "Release": "V3",
4730
  "Thinking": "Dense"
4731
  },
4732
  {
@@ -4737,7 +5287,7 @@
4737
  "Result": 45.96,
4738
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4739
  "Params": 14,
4740
- "Release": "V3",
4741
  "Thinking": "Dense"
4742
  },
4743
  {
@@ -4748,7 +5298,7 @@
4748
  "Result": 43.27,
4749
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4750
  "Params": 14,
4751
- "Release": "V3",
4752
  "Thinking": "Dense"
4753
  },
4754
  {
@@ -4759,7 +5309,7 @@
4759
  "Result": 51.01,
4760
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4761
  "Params": 14,
4762
- "Release": "V3",
4763
  "Thinking": "Dense"
4764
  },
4765
  {
@@ -4770,7 +5320,7 @@
4770
  "Result": 41.87,
4771
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4772
  "Params": 14,
4773
- "Release": "V3",
4774
  "Thinking": "Dense"
4775
  },
4776
  {
@@ -4781,7 +5331,7 @@
4781
  "Result": 46.23,
4782
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4783
  "Params": 14,
4784
- "Release": "V3",
4785
  "Thinking": "Dense"
4786
  },
4787
  {
@@ -4792,7 +5342,7 @@
4792
  "Result": 42.92,
4793
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4794
  "Params": 14,
4795
- "Release": "V3",
4796
  "Thinking": "Dense"
4797
  },
4798
  {
@@ -4803,7 +5353,7 @@
4803
  "Result": 48.33,
4804
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4805
  "Params": 14,
4806
- "Release": "V3",
4807
  "Thinking": "Dense"
4808
  },
4809
  {
@@ -4814,7 +5364,7 @@
4814
  "Result": 28.36,
4815
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4816
  "Params": 14,
4817
- "Release": "V3",
4818
  "Thinking": "Dense"
4819
  },
4820
  {
@@ -4825,7 +5375,7 @@
4825
  "Result": 54.23,
4826
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4827
  "Params": 14,
4828
- "Release": "V3",
4829
  "Thinking": "Dense"
4830
  },
4831
  {
@@ -4836,7 +5386,7 @@
4836
  "Result": 84.71,
4837
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4838
  "Params": 14,
4839
- "Release": "V3",
4840
  "Thinking": "Dense"
4841
  },
4842
  {
@@ -4847,7 +5397,7 @@
4847
  "Result": 27.69,
4848
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4849
  "Params": 14,
4850
- "Release": "V3",
4851
  "Thinking": "Dense"
4852
  },
4853
  {
@@ -4858,7 +5408,7 @@
4858
  "Result": 48.24,
4859
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4860
  "Params": 14,
4861
- "Release": "V3",
4862
  "Thinking": "Dense"
4863
  },
4864
  {
@@ -4869,7 +5419,7 @@
4869
  "Result": 27.69,
4870
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4871
  "Params": 14,
4872
- "Release": "V3",
4873
  "Thinking": "Dense"
4874
  },
4875
  {
@@ -4880,7 +5430,7 @@
4880
  "Result": 48.24,
4881
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4882
  "Params": 14,
4883
- "Release": "V3",
4884
  "Thinking": "Dense"
4885
  },
4886
  {
@@ -4891,7 +5441,7 @@
4891
  "Result": 27.36,
4892
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4893
  "Params": 14,
4894
- "Release": "V3",
4895
  "Thinking": "Dense"
4896
  },
4897
  {
@@ -4902,7 +5452,7 @@
4902
  "Result": 46.18,
4903
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4904
  "Params": 14,
4905
- "Release": "V3",
4906
  "Thinking": "Dense"
4907
  },
4908
  {
@@ -4913,7 +5463,7 @@
4913
  "Result": 27.34,
4914
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4915
  "Params": 14,
4916
- "Release": "V3",
4917
  "Thinking": "Dense"
4918
  },
4919
  {
@@ -4924,7 +5474,7 @@
4924
  "Result": 48.57,
4925
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4926
  "Params": 14,
4927
- "Release": "V3",
4928
  "Thinking": "Dense"
4929
  },
4930
  {
@@ -4935,7 +5485,7 @@
4935
  "Result": 26.2,
4936
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4937
  "Params": 14,
4938
- "Release": "V3",
4939
  "Thinking": "Dense"
4940
  },
4941
  {
@@ -4946,7 +5496,7 @@
4946
  "Result": 47.47,
4947
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4948
  "Params": 14,
4949
- "Release": "V3",
4950
  "Thinking": "Dense"
4951
  },
4952
  {
@@ -4957,7 +5507,7 @@
4957
  "Result": 70.9,
4958
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4959
  "Params": 8.2,
4960
- "Release": "V3",
4961
  "Thinking": "Reasoning"
4962
  },
4963
  {
@@ -4968,7 +5518,7 @@
4968
  "Result": 65.96,
4969
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4970
  "Params": 8.2,
4971
- "Release": "V3",
4972
  "Thinking": "Reasoning"
4973
  },
4974
  {
@@ -4979,7 +5529,7 @@
4979
  "Result": 51.03,
4980
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4981
  "Params": 8.2,
4982
- "Release": "V3",
4983
  "Thinking": "Reasoning"
4984
  },
4985
  {
@@ -4990,7 +5540,7 @@
4990
  "Result": 37.87,
4991
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4992
  "Params": 8.2,
4993
- "Release": "V3",
4994
  "Thinking": "Reasoning"
4995
  },
4996
  {
@@ -5001,7 +5551,7 @@
5001
  "Result": 51.03,
5002
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5003
  "Params": 8.2,
5004
- "Release": "V3",
5005
  "Thinking": "Reasoning"
5006
  },
5007
  {
@@ -5012,7 +5562,7 @@
5012
  "Result": 33.62,
5013
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5014
  "Params": 8.2,
5015
- "Release": "V3",
5016
  "Thinking": "Reasoning"
5017
  },
5018
  {
@@ -5023,7 +5573,7 @@
5023
  "Result": 50.57,
5024
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5025
  "Params": 8.2,
5026
- "Release": "V3",
5027
  "Thinking": "Reasoning"
5028
  },
5029
  {
@@ -5034,7 +5584,7 @@
5034
  "Result": 36.52,
5035
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5036
  "Params": 8.2,
5037
- "Release": "V3",
5038
  "Thinking": "Reasoning"
5039
  },
5040
  {
@@ -5045,7 +5595,7 @@
5045
  "Result": 48.9,
5046
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5047
  "Params": 8.2,
5048
- "Release": "V3",
5049
  "Thinking": "Reasoning"
5050
  },
5051
  {
@@ -5056,7 +5606,7 @@
5056
  "Result": 33.6,
5057
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5058
  "Params": 8.2,
5059
- "Release": "V3",
5060
  "Thinking": "Reasoning"
5061
  },
5062
  {
@@ -5067,7 +5617,7 @@
5067
  "Result": 49.93,
5068
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5069
  "Params": 8.2,
5070
- "Release": "V3",
5071
  "Thinking": "Reasoning"
5072
  },
5073
  {
@@ -5078,7 +5628,7 @@
5078
  "Result": 33.2,
5079
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5080
  "Params": 8.2,
5081
- "Release": "V3",
5082
  "Thinking": "Reasoning"
5083
  },
5084
  {
@@ -5089,7 +5639,7 @@
5089
  "Result": -1.0,
5090
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5091
  "Params": 8.2,
5092
- "Release": "V3",
5093
  "Thinking": "Reasoning"
5094
  },
5095
  {
@@ -5100,7 +5650,7 @@
5100
  "Result": 70.77,
5101
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5102
  "Params": 8.2,
5103
- "Release": "V3",
5104
  "Thinking": "Reasoning"
5105
  },
5106
  {
@@ -5111,7 +5661,7 @@
5111
  "Result": 83.53,
5112
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5113
  "Params": 8.2,
5114
- "Release": "V3",
5115
  "Thinking": "Reasoning"
5116
  },
5117
  {
@@ -5122,7 +5672,7 @@
5122
  "Result": 51.41,
5123
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5124
  "Params": 8.2,
5125
- "Release": "V3",
5126
  "Thinking": "Reasoning"
5127
  },
5128
  {
@@ -5133,7 +5683,7 @@
5133
  "Result": 40.0,
5134
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5135
  "Params": 8.2,
5136
- "Release": "V3",
5137
  "Thinking": "Reasoning"
5138
  },
5139
  {
@@ -5144,7 +5694,7 @@
5144
  "Result": 51.15,
5145
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5146
  "Params": 8.2,
5147
- "Release": "V3",
5148
  "Thinking": "Reasoning"
5149
  },
5150
  {
@@ -5155,7 +5705,7 @@
5155
  "Result": 38.82,
5156
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5157
  "Params": 8.2,
5158
- "Release": "V3",
5159
  "Thinking": "Reasoning"
5160
  },
5161
  {
@@ -5166,7 +5716,7 @@
5166
  "Result": 50.58,
5167
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5168
  "Params": 8.2,
5169
- "Release": "V3",
5170
  "Thinking": "Reasoning"
5171
  },
5172
  {
@@ -5177,7 +5727,7 @@
5177
  "Result": 37.47,
5178
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5179
  "Params": 8.2,
5180
- "Release": "V3",
5181
  "Thinking": "Reasoning"
5182
  },
5183
  {
@@ -5188,7 +5738,7 @@
5188
  "Result": 49.17,
5189
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5190
  "Params": 8.2,
5191
- "Release": "V3",
5192
  "Thinking": "Reasoning"
5193
  },
5194
  {
@@ -5199,7 +5749,7 @@
5199
  "Result": 38.76,
5200
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5201
  "Params": 8.2,
5202
- "Release": "V3",
5203
  "Thinking": "Reasoning"
5204
  },
5205
  {
@@ -5210,7 +5760,7 @@
5210
  "Result": 50.18,
5211
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5212
  "Params": 8.2,
5213
- "Release": "V3",
5214
  "Thinking": "Reasoning"
5215
  },
5216
  {
@@ -5221,7 +5771,7 @@
5221
  "Result": 38.73,
5222
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5223
  "Params": 8.2,
5224
- "Release": "V3",
5225
  "Thinking": "Reasoning"
5226
  },
5227
  {
 
1
  [
2
+ {
3
+ "Model": "DeepSeek V3.1 Terminus",
4
+ "Model Type": "General",
5
+ "Benchmark": "VerilogEval S2R",
6
+ "Task": "Syntax (STX)",
7
+ "Result": 95.77,
8
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
9
+ "Params": 685,
10
+ "Release": "V4",
11
+ "Thinking": "Reasoning"
12
+ },
13
+ {
14
+ "Model": "DeepSeek V3.1 Terminus",
15
+ "Model Type": "General",
16
+ "Benchmark": "RTLLM",
17
+ "Task": "Syntax (STX)",
18
+ "Result": 93.62,
19
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
20
+ "Params": 685,
21
+ "Release": "V4",
22
+ "Thinking": "Reasoning"
23
+ },
24
+ {
25
+ "Model": "DeepSeek V3.1 Terminus",
26
+ "Model Type": "General",
27
+ "Benchmark": "VerilogEval S2R",
28
+ "Task": "Functionality (FNC)",
29
+ "Result": 75.51,
30
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
31
+ "Params": 685,
32
+ "Release": "V4",
33
+ "Thinking": "Reasoning"
34
+ },
35
+ {
36
+ "Model": "DeepSeek V3.1 Terminus",
37
+ "Model Type": "General",
38
+ "Benchmark": "RTLLM",
39
+ "Task": "Functionality (FNC)",
40
+ "Result": 62.98,
41
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
42
+ "Params": 685,
43
+ "Release": "V4",
44
+ "Thinking": "Reasoning"
45
+ },
46
+ {
47
+ "Model": "DeepSeek V3.1 Terminus",
48
+ "Model Type": "General",
49
+ "Benchmark": "VerilogEval S2R",
50
+ "Task": "Synthesis (SYN)",
51
+ "Result": 75.51,
52
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
53
+ "Params": 685,
54
+ "Release": "V4",
55
+ "Thinking": "Reasoning"
56
+ },
57
+ {
58
+ "Model": "DeepSeek V3.1 Terminus",
59
+ "Model Type": "General",
60
+ "Benchmark": "RTLLM",
61
+ "Task": "Synthesis (SYN)",
62
+ "Result": 58.72,
63
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
64
+ "Params": 685,
65
+ "Release": "V4",
66
+ "Thinking": "Reasoning"
67
+ },
68
+ {
69
+ "Model": "DeepSeek V3.1 Terminus",
70
+ "Model Type": "General",
71
+ "Benchmark": "VerilogEval S2R",
72
+ "Task": "Power",
73
+ "Result": 75.51,
74
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
75
+ "Params": 685,
76
+ "Release": "V4",
77
+ "Thinking": "Reasoning"
78
+ },
79
+ {
80
+ "Model": "DeepSeek V3.1 Terminus",
81
+ "Model Type": "General",
82
+ "Benchmark": "RTLLM",
83
+ "Task": "Power",
84
+ "Result": 64.2,
85
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
86
+ "Params": 685,
87
+ "Release": "V4",
88
+ "Thinking": "Reasoning"
89
+ },
90
+ {
91
+ "Model": "DeepSeek V3.1 Terminus",
92
+ "Model Type": "General",
93
+ "Benchmark": "VerilogEval S2R",
94
+ "Task": "Performance",
95
+ "Result": 72.74,
96
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
97
+ "Params": 685,
98
+ "Release": "V4",
99
+ "Thinking": "Reasoning"
100
+ },
101
+ {
102
+ "Model": "DeepSeek V3.1 Terminus",
103
+ "Model Type": "General",
104
+ "Benchmark": "RTLLM",
105
+ "Task": "Performance",
106
+ "Result": 59.11,
107
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
108
+ "Params": 685,
109
+ "Release": "V4",
110
+ "Thinking": "Reasoning"
111
+ },
112
+ {
113
+ "Model": "DeepSeek V3.1 Terminus",
114
+ "Model Type": "General",
115
+ "Benchmark": "VerilogEval S2R",
116
+ "Task": "Area",
117
+ "Result": 74.42,
118
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
119
+ "Params": 685,
120
+ "Release": "V4",
121
+ "Thinking": "Reasoning"
122
+ },
123
+ {
124
+ "Model": "DeepSeek V3.1 Terminus",
125
+ "Model Type": "General",
126
+ "Benchmark": "RTLLM",
127
+ "Task": "Area",
128
+ "Result": 62.35,
129
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
130
+ "Params": 685,
131
+ "Release": "V4",
132
+ "Thinking": "Reasoning"
133
+ },
134
+ {
135
+ "Model": "DeepSeek V3.1 Terminus",
136
+ "Model Type": "General",
137
+ "Benchmark": "RTL-Repo",
138
+ "Task": "Exact Matching (EM)",
139
+ "Result": -1.0,
140
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
141
+ "Params": 685,
142
+ "Release": "V4",
143
+ "Thinking": "Reasoning"
144
+ },
145
+ {
146
+ "Model": "DeepSeek V3.1 Terminus",
147
+ "Model Type": "General",
148
+ "Benchmark": "VerilogEval MC",
149
+ "Task": "Syntax (STX)",
150
+ "Result": 95.13,
151
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
152
+ "Params": 685,
153
+ "Release": "V4",
154
+ "Thinking": "Reasoning"
155
+ },
156
+ {
157
+ "Model": "DeepSeek V3.1 Terminus",
158
+ "Model Type": "General",
159
+ "Benchmark": "VeriGen",
160
+ "Task": "Syntax (STX)",
161
+ "Result": 96.47,
162
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
163
+ "Params": 685,
164
+ "Release": "V4",
165
+ "Thinking": "Reasoning"
166
+ },
167
+ {
168
+ "Model": "DeepSeek V3.1 Terminus",
169
+ "Model Type": "General",
170
+ "Benchmark": "VerilogEval MC",
171
+ "Task": "Functionality (FNC)",
172
+ "Result": 77.05,
173
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
174
+ "Params": 685,
175
+ "Release": "V4",
176
+ "Thinking": "Reasoning"
177
+ },
178
+ {
179
+ "Model": "DeepSeek V3.1 Terminus",
180
+ "Model Type": "General",
181
+ "Benchmark": "VeriGen",
182
+ "Task": "Functionality (FNC)",
183
+ "Result": 78.82,
184
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
185
+ "Params": 685,
186
+ "Release": "V4",
187
+ "Thinking": "Reasoning"
188
+ },
189
+ {
190
+ "Model": "DeepSeek V3.1 Terminus",
191
+ "Model Type": "General",
192
+ "Benchmark": "VerilogEval MC",
193
+ "Task": "Synthesis (SYN)",
194
+ "Result": 76.79,
195
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
196
+ "Params": 685,
197
+ "Release": "V4",
198
+ "Thinking": "Reasoning"
199
+ },
200
+ {
201
+ "Model": "DeepSeek V3.1 Terminus",
202
+ "Model Type": "General",
203
+ "Benchmark": "VeriGen",
204
+ "Task": "Synthesis (SYN)",
205
+ "Result": 75.29,
206
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
207
+ "Params": 685,
208
+ "Release": "V4",
209
+ "Thinking": "Reasoning"
210
+ },
211
+ {
212
+ "Model": "DeepSeek V3.1 Terminus",
213
+ "Model Type": "General",
214
+ "Benchmark": "VerilogEval MC",
215
+ "Task": "Power",
216
+ "Result": 76.23,
217
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
218
+ "Params": 685,
219
+ "Release": "V4",
220
+ "Thinking": "Reasoning"
221
+ },
222
+ {
223
+ "Model": "DeepSeek V3.1 Terminus",
224
+ "Model Type": "General",
225
+ "Benchmark": "VeriGen",
226
+ "Task": "Power",
227
+ "Result": 69.87,
228
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
229
+ "Params": 685,
230
+ "Release": "V4",
231
+ "Thinking": "Reasoning"
232
+ },
233
+ {
234
+ "Model": "DeepSeek V3.1 Terminus",
235
+ "Model Type": "General",
236
+ "Benchmark": "VerilogEval MC",
237
+ "Task": "Performance",
238
+ "Result": 74.3,
239
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
240
+ "Params": 685,
241
+ "Release": "V4",
242
+ "Thinking": "Reasoning"
243
+ },
244
+ {
245
+ "Model": "DeepSeek V3.1 Terminus",
246
+ "Model Type": "General",
247
+ "Benchmark": "VeriGen",
248
+ "Task": "Performance",
249
+ "Result": 75.14,
250
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
251
+ "Params": 685,
252
+ "Release": "V4",
253
+ "Thinking": "Reasoning"
254
+ },
255
+ {
256
+ "Model": "DeepSeek V3.1 Terminus",
257
+ "Model Type": "General",
258
+ "Benchmark": "VerilogEval MC",
259
+ "Task": "Area",
260
+ "Result": 75.75,
261
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
262
+ "Params": 685,
263
+ "Release": "V4",
264
+ "Thinking": "Reasoning"
265
+ },
266
+ {
267
+ "Model": "DeepSeek V3.1 Terminus",
268
+ "Model Type": "General",
269
+ "Benchmark": "VeriGen",
270
+ "Task": "Area",
271
+ "Result": 77.59,
272
+ "Model URL": "https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus",
273
+ "Params": 685,
274
+ "Release": "V4",
275
+ "Thinking": "Reasoning"
276
+ },
277
+ {
278
+ "Model": "Kimi K2 Instruct 0905",
279
+ "Model Type": "General",
280
+ "Benchmark": "VerilogEval S2R",
281
+ "Task": "Syntax (STX)",
282
+ "Result": 96.28,
283
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
284
+ "Params": 1000,
285
+ "Release": "V4",
286
+ "Thinking": "Dense"
287
+ },
288
+ {
289
+ "Model": "Kimi K2 Instruct 0905",
290
+ "Model Type": "General",
291
+ "Benchmark": "RTLLM",
292
+ "Task": "Syntax (STX)",
293
+ "Result": 90.21,
294
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
295
+ "Params": 1000,
296
+ "Release": "V4",
297
+ "Thinking": "Dense"
298
+ },
299
+ {
300
+ "Model": "Kimi K2 Instruct 0905",
301
+ "Model Type": "General",
302
+ "Benchmark": "VerilogEval S2R",
303
+ "Task": "Functionality (FNC)",
304
+ "Result": 71.28,
305
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
306
+ "Params": 1000,
307
+ "Release": "V4",
308
+ "Thinking": "Dense"
309
+ },
310
+ {
311
+ "Model": "Kimi K2 Instruct 0905",
312
+ "Model Type": "General",
313
+ "Benchmark": "RTLLM",
314
+ "Task": "Functionality (FNC)",
315
+ "Result": 64.68,
316
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
317
+ "Params": 1000,
318
+ "Release": "V4",
319
+ "Thinking": "Dense"
320
+ },
321
+ {
322
+ "Model": "Kimi K2 Instruct 0905",
323
+ "Model Type": "General",
324
+ "Benchmark": "VerilogEval S2R",
325
+ "Task": "Synthesis (SYN)",
326
+ "Result": 70.77,
327
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
328
+ "Params": 1000,
329
+ "Release": "V4",
330
+ "Thinking": "Dense"
331
+ },
332
+ {
333
+ "Model": "Kimi K2 Instruct 0905",
334
+ "Model Type": "General",
335
+ "Benchmark": "RTLLM",
336
+ "Task": "Synthesis (SYN)",
337
+ "Result": 60.43,
338
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
339
+ "Params": 1000,
340
+ "Release": "V4",
341
+ "Thinking": "Dense"
342
+ },
343
+ {
344
+ "Model": "Kimi K2 Instruct 0905",
345
+ "Model Type": "General",
346
+ "Benchmark": "VerilogEval S2R",
347
+ "Task": "Power",
348
+ "Result": 70.07,
349
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
350
+ "Params": 1000,
351
+ "Release": "V4",
352
+ "Thinking": "Dense"
353
+ },
354
+ {
355
+ "Model": "Kimi K2 Instruct 0905",
356
+ "Model Type": "General",
357
+ "Benchmark": "RTLLM",
358
+ "Task": "Power",
359
+ "Result": 62.94,
360
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
361
+ "Params": 1000,
362
+ "Release": "V4",
363
+ "Thinking": "Dense"
364
+ },
365
+ {
366
+ "Model": "Kimi K2 Instruct 0905",
367
+ "Model Type": "General",
368
+ "Benchmark": "VerilogEval S2R",
369
+ "Task": "Performance",
370
+ "Result": 68.77,
371
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
372
+ "Params": 1000,
373
+ "Release": "V4",
374
+ "Thinking": "Dense"
375
+ },
376
+ {
377
+ "Model": "Kimi K2 Instruct 0905",
378
+ "Model Type": "General",
379
+ "Benchmark": "RTLLM",
380
+ "Task": "Performance",
381
+ "Result": 58.35,
382
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
383
+ "Params": 1000,
384
+ "Release": "V4",
385
+ "Thinking": "Dense"
386
+ },
387
+ {
388
+ "Model": "Kimi K2 Instruct 0905",
389
+ "Model Type": "General",
390
+ "Benchmark": "VerilogEval S2R",
391
+ "Task": "Area",
392
+ "Result": 70.32,
393
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
394
+ "Params": 1000,
395
+ "Release": "V4",
396
+ "Thinking": "Dense"
397
+ },
398
+ {
399
+ "Model": "Kimi K2 Instruct 0905",
400
+ "Model Type": "General",
401
+ "Benchmark": "RTLLM",
402
+ "Task": "Area",
403
+ "Result": 63.08,
404
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
405
+ "Params": 1000,
406
+ "Release": "V4",
407
+ "Thinking": "Dense"
408
+ },
409
+ {
410
+ "Model": "Kimi K2 Instruct 0905",
411
+ "Model Type": "General",
412
+ "Benchmark": "RTL-Repo",
413
+ "Task": "Exact Matching (EM)",
414
+ "Result": 33.65,
415
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
416
+ "Params": 1000,
417
+ "Release": "V4",
418
+ "Thinking": "Dense"
419
+ },
420
+ {
421
+ "Model": "Kimi K2 Instruct 0905",
422
+ "Model Type": "General",
423
+ "Benchmark": "VerilogEval MC",
424
+ "Task": "Syntax (STX)",
425
+ "Result": 98.59,
426
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
427
+ "Params": 1000,
428
+ "Release": "V4",
429
+ "Thinking": "Dense"
430
+ },
431
+ {
432
+ "Model": "Kimi K2 Instruct 0905",
433
+ "Model Type": "General",
434
+ "Benchmark": "VeriGen",
435
+ "Task": "Syntax (STX)",
436
+ "Result": 95.29,
437
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
438
+ "Params": 1000,
439
+ "Release": "V4",
440
+ "Thinking": "Dense"
441
+ },
442
+ {
443
+ "Model": "Kimi K2 Instruct 0905",
444
+ "Model Type": "General",
445
+ "Benchmark": "VerilogEval MC",
446
+ "Task": "Functionality (FNC)",
447
+ "Result": 74.74,
448
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
449
+ "Params": 1000,
450
+ "Release": "V4",
451
+ "Thinking": "Dense"
452
+ },
453
+ {
454
+ "Model": "Kimi K2 Instruct 0905",
455
+ "Model Type": "General",
456
+ "Benchmark": "VeriGen",
457
+ "Task": "Functionality (FNC)",
458
+ "Result": 69.41,
459
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
460
+ "Params": 1000,
461
+ "Release": "V4",
462
+ "Thinking": "Dense"
463
+ },
464
+ {
465
+ "Model": "Kimi K2 Instruct 0905",
466
+ "Model Type": "General",
467
+ "Benchmark": "VerilogEval MC",
468
+ "Task": "Synthesis (SYN)",
469
+ "Result": 74.1,
470
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
471
+ "Params": 1000,
472
+ "Release": "V4",
473
+ "Thinking": "Dense"
474
+ },
475
+ {
476
+ "Model": "Kimi K2 Instruct 0905",
477
+ "Model Type": "General",
478
+ "Benchmark": "VeriGen",
479
+ "Task": "Synthesis (SYN)",
480
+ "Result": 69.41,
481
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
482
+ "Params": 1000,
483
+ "Release": "V4",
484
+ "Thinking": "Dense"
485
+ },
486
+ {
487
+ "Model": "Kimi K2 Instruct 0905",
488
+ "Model Type": "General",
489
+ "Benchmark": "VerilogEval MC",
490
+ "Task": "Power",
491
+ "Result": 73.29,
492
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
493
+ "Params": 1000,
494
+ "Release": "V4",
495
+ "Thinking": "Dense"
496
+ },
497
+ {
498
+ "Model": "Kimi K2 Instruct 0905",
499
+ "Model Type": "General",
500
+ "Benchmark": "VeriGen",
501
+ "Task": "Power",
502
+ "Result": 57.47,
503
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
504
+ "Params": 1000,
505
+ "Release": "V4",
506
+ "Thinking": "Dense"
507
+ },
508
+ {
509
+ "Model": "Kimi K2 Instruct 0905",
510
+ "Model Type": "General",
511
+ "Benchmark": "VerilogEval MC",
512
+ "Task": "Performance",
513
+ "Result": 71.1,
514
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
515
+ "Params": 1000,
516
+ "Release": "V4",
517
+ "Thinking": "Dense"
518
+ },
519
+ {
520
+ "Model": "Kimi K2 Instruct 0905",
521
+ "Model Type": "General",
522
+ "Benchmark": "VeriGen",
523
+ "Task": "Performance",
524
+ "Result": 69.9,
525
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
526
+ "Params": 1000,
527
+ "Release": "V4",
528
+ "Thinking": "Dense"
529
+ },
530
+ {
531
+ "Model": "Kimi K2 Instruct 0905",
532
+ "Model Type": "General",
533
+ "Benchmark": "VerilogEval MC",
534
+ "Task": "Area",
535
+ "Result": 73.0,
536
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
537
+ "Params": 1000,
538
+ "Release": "V4",
539
+ "Thinking": "Dense"
540
+ },
541
+ {
542
+ "Model": "Kimi K2 Instruct 0905",
543
+ "Model Type": "General",
544
+ "Benchmark": "VeriGen",
545
+ "Task": "Area",
546
+ "Result": 69.51,
547
+ "Model URL": "https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905",
548
+ "Params": 1000,
549
+ "Release": "V4",
550
+ "Thinking": "Dense"
551
+ },
552
  {
553
  "Model": "Gemini 2.5 Flash (Medium)",
554
  "Model Type": "General",
 
557
  "Result": 91.54,
558
  "Model URL": "https://huggingface.co/google",
559
  "Params": null,
560
+ "Release": "V4",
561
  "Thinking": "Reasoning"
562
  },
563
  {
 
568
  "Result": 81.7,
569
  "Model URL": "https://huggingface.co/google",
570
  "Params": null,
571
+ "Release": "V4",
572
  "Thinking": "Reasoning"
573
  },
574
  {
 
579
  "Result": 69.74,
580
  "Model URL": "https://huggingface.co/google",
581
  "Params": null,
582
+ "Release": "V4",
583
  "Thinking": "Reasoning"
584
  },
585
  {
 
590
  "Result": 64.68,
591
  "Model URL": "https://huggingface.co/google",
592
  "Params": null,
593
+ "Release": "V4",
594
  "Thinking": "Reasoning"
595
  },
596
  {
 
601
  "Result": 69.23,
602
  "Model URL": "https://huggingface.co/google",
603
  "Params": null,
604
+ "Release": "V4",
605
  "Thinking": "Reasoning"
606
  },
607
  {
 
612
  "Result": 46.38,
613
  "Model URL": "https://huggingface.co/google",
614
  "Params": null,
615
+ "Release": "V4",
616
  "Thinking": "Reasoning"
617
  },
618
  {
 
623
  "Result": 68.4,
624
  "Model URL": "https://huggingface.co/google",
625
  "Params": null,
626
+ "Release": "V4",
627
  "Thinking": "Reasoning"
628
  },
629
  {
 
634
  "Result": 51.11,
635
  "Model URL": "https://huggingface.co/google",
636
  "Params": null,
637
+ "Release": "V4",
638
  "Thinking": "Reasoning"
639
  },
640
  {
 
645
  "Result": 66.57,
646
  "Model URL": "https://huggingface.co/google",
647
  "Params": null,
648
+ "Release": "V4",
649
  "Thinking": "Reasoning"
650
  },
651
  {
 
656
  "Result": 45.1,
657
  "Model URL": "https://huggingface.co/google",
658
  "Params": null,
659
+ "Release": "V4",
660
  "Thinking": "Reasoning"
661
  },
662
  {
 
667
  "Result": 68.27,
668
  "Model URL": "https://huggingface.co/google",
669
  "Params": null,
670
+ "Release": "V4",
671
  "Thinking": "Reasoning"
672
  },
673
  {
 
678
  "Result": 49.25,
679
  "Model URL": "https://huggingface.co/google",
680
  "Params": null,
681
+ "Release": "V4",
682
  "Thinking": "Reasoning"
683
  },
684
  {
 
689
  "Result": -1.0,
690
  "Model URL": "https://huggingface.co/google",
691
  "Params": null,
692
+ "Release": "V4",
693
  "Thinking": "Reasoning"
694
  },
695
  {
 
700
  "Result": 90.64,
701
  "Model URL": "https://huggingface.co/google",
702
  "Params": null,
703
+ "Release": "V4",
704
  "Thinking": "Reasoning"
705
  },
706
  {
 
711
  "Result": 96.47,
712
  "Model URL": "https://huggingface.co/google",
713
  "Params": null,
714
+ "Release": "V4",
715
  "Thinking": "Reasoning"
716
  },
717
  {
 
722
  "Result": 71.67,
723
  "Model URL": "https://huggingface.co/google",
724
  "Params": null,
725
+ "Release": "V4",
726
  "Thinking": "Reasoning"
727
  },
728
  {
 
733
  "Result": 77.65,
734
  "Model URL": "https://huggingface.co/google",
735
  "Params": null,
736
+ "Release": "V4",
737
  "Thinking": "Reasoning"
738
  },
739
  {
 
744
  "Result": 71.28,
745
  "Model URL": "https://huggingface.co/google",
746
  "Params": null,
747
+ "Release": "V4",
748
  "Thinking": "Reasoning"
749
  },
750
  {
 
755
  "Result": 77.65,
756
  "Model URL": "https://huggingface.co/google",
757
  "Params": null,
758
+ "Release": "V4",
759
  "Thinking": "Reasoning"
760
  },
761
  {
 
766
  "Result": 70.28,
767
  "Model URL": "https://huggingface.co/google",
768
  "Params": null,
769
+ "Release": "V4",
770
  "Thinking": "Reasoning"
771
  },
772
  {
 
777
  "Result": 71.27,
778
  "Model URL": "https://huggingface.co/google",
779
  "Params": null,
780
+ "Release": "V4",
781
  "Thinking": "Reasoning"
782
  },
783
  {
 
788
  "Result": 68.06,
789
  "Model URL": "https://huggingface.co/google",
790
  "Params": null,
791
+ "Release": "V4",
792
  "Thinking": "Reasoning"
793
  },
794
  {
 
799
  "Result": 77.35,
800
  "Model URL": "https://huggingface.co/google",
801
  "Params": null,
802
+ "Release": "V4",
803
  "Thinking": "Reasoning"
804
  },
805
  {
 
810
  "Result": 70.32,
811
  "Model URL": "https://huggingface.co/google",
812
  "Params": null,
813
+ "Release": "V4",
814
  "Thinking": "Reasoning"
815
  },
816
  {
 
821
  "Result": 79.45,
822
  "Model URL": "https://huggingface.co/google",
823
  "Params": null,
824
+ "Release": "V4",
825
  "Thinking": "Reasoning"
826
  },
827
  {
 
3032
  "Result": 89.36,
3033
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3034
  "Params": 36.2,
3035
+ "Release": "V4",
3036
  "Thinking": "Reasoning"
3037
  },
3038
  {
 
3043
  "Result": 83.4,
3044
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3045
  "Params": 36.2,
3046
+ "Release": "V4",
3047
  "Thinking": "Reasoning"
3048
  },
3049
  {
 
3054
  "Result": 71.15,
3055
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3056
  "Params": 36.2,
3057
+ "Release": "V4",
3058
  "Thinking": "Reasoning"
3059
  },
3060
  {
 
3065
  "Result": 61.7,
3066
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3067
  "Params": 36.2,
3068
+ "Release": "V4",
3069
  "Thinking": "Reasoning"
3070
  },
3071
  {
 
3076
  "Result": 70.9,
3077
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3078
  "Params": 36.2,
3079
+ "Release": "V4",
3080
  "Thinking": "Reasoning"
3081
  },
3082
  {
 
3087
  "Result": 57.02,
3088
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3089
  "Params": 36.2,
3090
+ "Release": "V4",
3091
  "Thinking": "Reasoning"
3092
  },
3093
  {
 
3098
  "Result": 70.53,
3099
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3100
  "Params": 36.2,
3101
+ "Release": "V4",
3102
  "Thinking": "Reasoning"
3103
  },
3104
  {
 
3109
  "Result": 65.33,
3110
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3111
  "Params": 36.2,
3112
+ "Release": "V4",
3113
  "Thinking": "Reasoning"
3114
  },
3115
  {
 
3120
  "Result": 68.22,
3121
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3122
  "Params": 36.2,
3123
+ "Release": "V4",
3124
  "Thinking": "Reasoning"
3125
  },
3126
  {
 
3131
  "Result": 57.32,
3132
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3133
  "Params": 36.2,
3134
+ "Release": "V4",
3135
  "Thinking": "Reasoning"
3136
  },
3137
  {
 
3142
  "Result": 70.33,
3143
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3144
  "Params": 36.2,
3145
+ "Release": "V4",
3146
  "Thinking": "Reasoning"
3147
  },
3148
  {
 
3153
  "Result": 61.39,
3154
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3155
  "Params": 36.2,
3156
+ "Release": "V4",
3157
  "Thinking": "Reasoning"
3158
  },
3159
  {
 
3164
  "Result": -1.0,
3165
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3166
  "Params": 36.2,
3167
+ "Release": "V4",
3168
  "Thinking": "Reasoning"
3169
  },
3170
  {
 
3175
  "Result": 91.03,
3176
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3177
  "Params": 36.2,
3178
+ "Release": "V4",
3179
  "Thinking": "Reasoning"
3180
  },
3181
  {
 
3186
  "Result": 91.76,
3187
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3188
  "Params": 36.2,
3189
+ "Release": "V4",
3190
  "Thinking": "Reasoning"
3191
  },
3192
  {
 
3197
  "Result": 73.85,
3198
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3199
  "Params": 36.2,
3200
+ "Release": "V4",
3201
  "Thinking": "Reasoning"
3202
  },
3203
  {
 
3208
  "Result": 70.59,
3209
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3210
  "Params": 36.2,
3211
+ "Release": "V4",
3212
  "Thinking": "Reasoning"
3213
  },
3214
  {
 
3219
  "Result": 73.46,
3220
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3221
  "Params": 36.2,
3222
+ "Release": "V4",
3223
  "Thinking": "Reasoning"
3224
  },
3225
  {
 
3230
  "Result": 70.59,
3231
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3232
  "Params": 36.2,
3233
+ "Release": "V4",
3234
  "Thinking": "Reasoning"
3235
  },
3236
  {
 
3241
  "Result": 73.04,
3242
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3243
  "Params": 36.2,
3244
+ "Release": "V4",
3245
  "Thinking": "Reasoning"
3246
  },
3247
  {
 
3252
  "Result": 66.11,
3253
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3254
  "Params": 36.2,
3255
+ "Release": "V4",
3256
  "Thinking": "Reasoning"
3257
  },
3258
  {
 
3263
  "Result": 70.89,
3264
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3265
  "Params": 36.2,
3266
+ "Release": "V4",
3267
  "Thinking": "Reasoning"
3268
  },
3269
  {
 
3274
  "Result": 70.19,
3275
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3276
  "Params": 36.2,
3277
+ "Release": "V4",
3278
  "Thinking": "Reasoning"
3279
  },
3280
  {
 
3285
  "Result": 72.6,
3286
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3287
  "Params": 36.2,
3288
+ "Release": "V4",
3289
  "Thinking": "Reasoning"
3290
  },
3291
  {
 
3296
  "Result": 72.45,
3297
  "Model URL": "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct",
3298
  "Params": 36.2,
3299
+ "Release": "V4",
3300
  "Thinking": "Reasoning"
3301
  },
3302
  {
 
4957
  "Result": 80.64,
4958
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4959
  "Params": 14,
4960
+ "Release": "V4",
4961
  "Thinking": "Reasoning"
4962
  },
4963
  {
 
4968
  "Result": 79.57,
4969
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4970
  "Params": 14,
4971
+ "Release": "V4",
4972
  "Thinking": "Reasoning"
4973
  },
4974
  {
 
4979
  "Result": 56.15,
4980
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4981
  "Params": 14,
4982
+ "Release": "V4",
4983
  "Thinking": "Reasoning"
4984
  },
4985
  {
 
4990
  "Result": 41.28,
4991
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4992
  "Params": 14,
4993
+ "Release": "V4",
4994
  "Thinking": "Reasoning"
4995
  },
4996
  {
 
5001
  "Result": 55.77,
5002
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5003
  "Params": 14,
5004
+ "Release": "V4",
5005
  "Thinking": "Reasoning"
5006
  },
5007
  {
 
5012
  "Result": 34.47,
5013
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5014
  "Params": 14,
5015
+ "Release": "V4",
5016
  "Thinking": "Reasoning"
5017
  },
5018
  {
 
5023
  "Result": 55.85,
5024
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5025
  "Params": 14,
5026
+ "Release": "V4",
5027
  "Thinking": "Reasoning"
5028
  },
5029
  {
 
5034
  "Result": 38.35,
5035
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5036
  "Params": 14,
5037
+ "Release": "V4",
5038
  "Thinking": "Reasoning"
5039
  },
5040
  {
 
5045
  "Result": 53.71,
5046
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5047
  "Params": 14,
5048
+ "Release": "V4",
5049
  "Thinking": "Reasoning"
5050
  },
5051
  {
 
5056
  "Result": 32.73,
5057
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5058
  "Params": 14,
5059
+ "Release": "V4",
5060
  "Thinking": "Reasoning"
5061
  },
5062
  {
 
5067
  "Result": 55.65,
5068
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5069
  "Params": 14,
5070
+ "Release": "V4",
5071
  "Thinking": "Reasoning"
5072
  },
5073
  {
 
5078
  "Result": 36.13,
5079
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5080
  "Params": 14,
5081
+ "Release": "V4",
5082
  "Thinking": "Reasoning"
5083
  },
5084
  {
 
5089
  "Result": -1.0,
5090
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5091
  "Params": 14,
5092
+ "Release": "V4",
5093
  "Thinking": "Reasoning"
5094
  },
5095
  {
 
5100
  "Result": 52.18,
5101
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5102
  "Params": 14,
5103
+ "Release": "V4",
5104
  "Thinking": "Reasoning"
5105
  },
5106
  {
 
5111
  "Result": 84.71,
5112
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5113
  "Params": 14,
5114
+ "Release": "V4",
5115
  "Thinking": "Reasoning"
5116
  },
5117
  {
 
5122
  "Result": 36.92,
5123
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5124
  "Params": 14,
5125
+ "Release": "V4",
5126
  "Thinking": "Reasoning"
5127
  },
5128
  {
 
5133
  "Result": 60.0,
5134
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5135
  "Params": 14,
5136
+ "Release": "V4",
5137
  "Thinking": "Reasoning"
5138
  },
5139
  {
 
5144
  "Result": 36.67,
5145
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5146
  "Params": 14,
5147
+ "Release": "V4",
5148
  "Thinking": "Reasoning"
5149
  },
5150
  {
 
5155
  "Result": 58.82,
5156
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5157
  "Params": 14,
5158
+ "Release": "V4",
5159
  "Thinking": "Reasoning"
5160
  },
5161
  {
 
5166
  "Result": 36.21,
5167
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5168
  "Params": 14,
5169
+ "Release": "V4",
5170
  "Thinking": "Reasoning"
5171
  },
5172
  {
 
5177
  "Result": 52.97,
5178
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5179
  "Params": 14,
5180
+ "Release": "V4",
5181
  "Thinking": "Reasoning"
5182
  },
5183
  {
 
5188
  "Result": 36.02,
5189
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5190
  "Params": 14,
5191
+ "Release": "V4",
5192
  "Thinking": "Reasoning"
5193
  },
5194
  {
 
5199
  "Result": 58.57,
5200
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5201
  "Params": 14,
5202
+ "Release": "V4",
5203
  "Thinking": "Reasoning"
5204
  },
5205
  {
 
5210
  "Result": 36.12,
5211
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5212
  "Params": 14,
5213
+ "Release": "V4",
5214
  "Thinking": "Reasoning"
5215
  },
5216
  {
 
5221
  "Result": 59.8,
5222
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5223
  "Params": 14,
5224
+ "Release": "V4",
5225
  "Thinking": "Reasoning"
5226
  },
5227
  {
 
5232
  "Result": 72.18,
5233
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5234
  "Params": 14,
5235
+ "Release": "V4",
5236
  "Thinking": "Dense"
5237
  },
5238
  {
 
5243
  "Result": 85.11,
5244
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5245
  "Params": 14,
5246
+ "Release": "V4",
5247
  "Thinking": "Dense"
5248
  },
5249
  {
 
5254
  "Result": 43.85,
5255
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5256
  "Params": 14,
5257
+ "Release": "V4",
5258
  "Thinking": "Dense"
5259
  },
5260
  {
 
5265
  "Result": 50.64,
5266
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5267
  "Params": 14,
5268
+ "Release": "V4",
5269
  "Thinking": "Dense"
5270
  },
5271
  {
 
5276
  "Result": 43.85,
5277
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5278
  "Params": 14,
5279
+ "Release": "V4",
5280
  "Thinking": "Dense"
5281
  },
5282
  {
 
5287
  "Result": 45.96,
5288
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5289
  "Params": 14,
5290
+ "Release": "V4",
5291
  "Thinking": "Dense"
5292
  },
5293
  {
 
5298
  "Result": 43.27,
5299
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5300
  "Params": 14,
5301
+ "Release": "V4",
5302
  "Thinking": "Dense"
5303
  },
5304
  {
 
5309
  "Result": 51.01,
5310
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5311
  "Params": 14,
5312
+ "Release": "V4",
5313
  "Thinking": "Dense"
5314
  },
5315
  {
 
5320
  "Result": 41.87,
5321
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5322
  "Params": 14,
5323
+ "Release": "V4",
5324
  "Thinking": "Dense"
5325
  },
5326
  {
 
5331
  "Result": 46.23,
5332
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5333
  "Params": 14,
5334
+ "Release": "V4",
5335
  "Thinking": "Dense"
5336
  },
5337
  {
 
5342
  "Result": 42.92,
5343
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5344
  "Params": 14,
5345
+ "Release": "V4",
5346
  "Thinking": "Dense"
5347
  },
5348
  {
 
5353
  "Result": 48.33,
5354
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5355
  "Params": 14,
5356
+ "Release": "V4",
5357
  "Thinking": "Dense"
5358
  },
5359
  {
 
5364
  "Result": 28.36,
5365
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5366
  "Params": 14,
5367
+ "Release": "V4",
5368
  "Thinking": "Dense"
5369
  },
5370
  {
 
5375
  "Result": 54.23,
5376
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5377
  "Params": 14,
5378
+ "Release": "V4",
5379
  "Thinking": "Dense"
5380
  },
5381
  {
 
5386
  "Result": 84.71,
5387
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5388
  "Params": 14,
5389
+ "Release": "V4",
5390
  "Thinking": "Dense"
5391
  },
5392
  {
 
5397
  "Result": 27.69,
5398
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5399
  "Params": 14,
5400
+ "Release": "V4",
5401
  "Thinking": "Dense"
5402
  },
5403
  {
 
5408
  "Result": 48.24,
5409
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5410
  "Params": 14,
5411
+ "Release": "V4",
5412
  "Thinking": "Dense"
5413
  },
5414
  {
 
5419
  "Result": 27.69,
5420
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5421
  "Params": 14,
5422
+ "Release": "V4",
5423
  "Thinking": "Dense"
5424
  },
5425
  {
 
5430
  "Result": 48.24,
5431
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5432
  "Params": 14,
5433
+ "Release": "V4",
5434
  "Thinking": "Dense"
5435
  },
5436
  {
 
5441
  "Result": 27.36,
5442
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5443
  "Params": 14,
5444
+ "Release": "V4",
5445
  "Thinking": "Dense"
5446
  },
5447
  {
 
5452
  "Result": 46.18,
5453
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5454
  "Params": 14,
5455
+ "Release": "V4",
5456
  "Thinking": "Dense"
5457
  },
5458
  {
 
5463
  "Result": 27.34,
5464
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5465
  "Params": 14,
5466
+ "Release": "V4",
5467
  "Thinking": "Dense"
5468
  },
5469
  {
 
5474
  "Result": 48.57,
5475
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5476
  "Params": 14,
5477
+ "Release": "V4",
5478
  "Thinking": "Dense"
5479
  },
5480
  {
 
5485
  "Result": 26.2,
5486
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5487
  "Params": 14,
5488
+ "Release": "V4",
5489
  "Thinking": "Dense"
5490
  },
5491
  {
 
5496
  "Result": 47.47,
5497
  "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
5498
  "Params": 14,
5499
+ "Release": "V4",
5500
  "Thinking": "Dense"
5501
  },
5502
  {
 
5507
  "Result": 70.9,
5508
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5509
  "Params": 8.2,
5510
+ "Release": "V4",
5511
  "Thinking": "Reasoning"
5512
  },
5513
  {
 
5518
  "Result": 65.96,
5519
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5520
  "Params": 8.2,
5521
+ "Release": "V4",
5522
  "Thinking": "Reasoning"
5523
  },
5524
  {
 
5529
  "Result": 51.03,
5530
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5531
  "Params": 8.2,
5532
+ "Release": "V4",
5533
  "Thinking": "Reasoning"
5534
  },
5535
  {
 
5540
  "Result": 37.87,
5541
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5542
  "Params": 8.2,
5543
+ "Release": "V4",
5544
  "Thinking": "Reasoning"
5545
  },
5546
  {
 
5551
  "Result": 51.03,
5552
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5553
  "Params": 8.2,
5554
+ "Release": "V4",
5555
  "Thinking": "Reasoning"
5556
  },
5557
  {
 
5562
  "Result": 33.62,
5563
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5564
  "Params": 8.2,
5565
+ "Release": "V4",
5566
  "Thinking": "Reasoning"
5567
  },
5568
  {
 
5573
  "Result": 50.57,
5574
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5575
  "Params": 8.2,
5576
+ "Release": "V4",
5577
  "Thinking": "Reasoning"
5578
  },
5579
  {
 
5584
  "Result": 36.52,
5585
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5586
  "Params": 8.2,
5587
+ "Release": "V4",
5588
  "Thinking": "Reasoning"
5589
  },
5590
  {
 
5595
  "Result": 48.9,
5596
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5597
  "Params": 8.2,
5598
+ "Release": "V4",
5599
  "Thinking": "Reasoning"
5600
  },
5601
  {
 
5606
  "Result": 33.6,
5607
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5608
  "Params": 8.2,
5609
+ "Release": "V4",
5610
  "Thinking": "Reasoning"
5611
  },
5612
  {
 
5617
  "Result": 49.93,
5618
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5619
  "Params": 8.2,
5620
+ "Release": "V4",
5621
  "Thinking": "Reasoning"
5622
  },
5623
  {
 
5628
  "Result": 33.2,
5629
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5630
  "Params": 8.2,
5631
+ "Release": "V4",
5632
  "Thinking": "Reasoning"
5633
  },
5634
  {
 
5639
  "Result": -1.0,
5640
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5641
  "Params": 8.2,
5642
+ "Release": "V4",
5643
  "Thinking": "Reasoning"
5644
  },
5645
  {
 
5650
  "Result": 70.77,
5651
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5652
  "Params": 8.2,
5653
+ "Release": "V4",
5654
  "Thinking": "Reasoning"
5655
  },
5656
  {
 
5661
  "Result": 83.53,
5662
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5663
  "Params": 8.2,
5664
+ "Release": "V4",
5665
  "Thinking": "Reasoning"
5666
  },
5667
  {
 
5672
  "Result": 51.41,
5673
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5674
  "Params": 8.2,
5675
+ "Release": "V4",
5676
  "Thinking": "Reasoning"
5677
  },
5678
  {
 
5683
  "Result": 40.0,
5684
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5685
  "Params": 8.2,
5686
+ "Release": "V4",
5687
  "Thinking": "Reasoning"
5688
  },
5689
  {
 
5694
  "Result": 51.15,
5695
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5696
  "Params": 8.2,
5697
+ "Release": "V4",
5698
  "Thinking": "Reasoning"
5699
  },
5700
  {
 
5705
  "Result": 38.82,
5706
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5707
  "Params": 8.2,
5708
+ "Release": "V4",
5709
  "Thinking": "Reasoning"
5710
  },
5711
  {
 
5716
  "Result": 50.58,
5717
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5718
  "Params": 8.2,
5719
+ "Release": "V4",
5720
  "Thinking": "Reasoning"
5721
  },
5722
  {
 
5727
  "Result": 37.47,
5728
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5729
  "Params": 8.2,
5730
+ "Release": "V4",
5731
  "Thinking": "Reasoning"
5732
  },
5733
  {
 
5738
  "Result": 49.17,
5739
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5740
  "Params": 8.2,
5741
+ "Release": "V4",
5742
  "Thinking": "Reasoning"
5743
  },
5744
  {
 
5749
  "Result": 38.76,
5750
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5751
  "Params": 8.2,
5752
+ "Release": "V4",
5753
  "Thinking": "Reasoning"
5754
  },
5755
  {
 
5760
  "Result": 50.18,
5761
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5762
  "Params": 8.2,
5763
+ "Release": "V4",
5764
  "Thinking": "Reasoning"
5765
  },
5766
  {
 
5771
  "Result": 38.73,
5772
  "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
5773
  "Params": 8.2,
5774
+ "Release": "V4",
5775
  "Thinking": "Reasoning"
5776
  },
5777
  {
results/results_verilator_november_2025.csv CHANGED
@@ -1,5 +1,7 @@
1
  ,Syntax (STX),Syntax (STX),Functionality (FNC),Functionality (FNC),Synthesis (SYN),Synthesis (SYN),Power,Power,Performance,Performance,Area,Area,EM,Syntax (STX),Syntax (STX),Functionality (FNC),Functionality (FNC),Synthesis (SYN),Synthesis (SYN),Power,Power,Performance,Performance,Area,Area
2
  ,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,RTL-Repo,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen
 
 
3
  Gemini 2.5 Flash (Medium),91.540000000000006,81.700000000000003,69.739999999999995,64.680000000000007,69.230000000000004,46.380000000000003,68.400000000000006,51.109999999999999,66.569999999999993,45.100000000000001,68.269999999999996,49.25,-1,90.640000000000001,96.469999999999999,71.670000000000002,77.650000000000006,71.280000000000001,77.650000000000006,70.280000000000001,71.269999999999996,68.060000000000002,77.349999999999994,70.319999999999993,79.450000000000003
4
  DeepSeek R1-0528,96.540000000000006,89.790000000000006,79.099999999999994,67.659999999999997,78.969999999999999,64.260000000000005,78.439999999999998,73.319999999999993,76.290000000000006,65.379999999999995,78.200000000000003,70.859999999999999,-1,94.739999999999995,98.819999999999993,80,80,79.870000000000005,74.120000000000005,79.659999999999997,63.859999999999999,77.280000000000001,74.049999999999997,79.409999999999997,77.010000000000005
5
  DeepSeek R1,97.689999999999998,93.189999999999998,79.359999999999999,67.659999999999997,79.099999999999994,62.979999999999997,79.010000000000005,71.379999999999995,76.489999999999995,63.740000000000002,78.629999999999995,69.819999999999993,-1,97.439999999999998,95.290000000000006,77.819999999999993,74.120000000000005,77.689999999999998,74.120000000000005,77.299999999999997,64.579999999999998,75.030000000000001,74.420000000000002,76.939999999999998,77.150000000000006
@@ -65,5 +67,4 @@ CodeV-DS-6.7B,34.359999999999999,69.790000000000006,14.74,41.280000000000001,14.
65
  ,,,,,,,,,,,,,,,,,,,,,,,,,
66
  ,,,,,,,,,,,,,,,,,,,,,,,,,
67
  ,,,,,,,,,,,,,,,,,,,,,,,,,
68
- ,,,,,,,,,,,,,,,,,,,,,,,,,
69
 
 
1
  ,Syntax (STX),Syntax (STX),Functionality (FNC),Functionality (FNC),Synthesis (SYN),Synthesis (SYN),Power,Power,Performance,Performance,Area,Area,EM,Syntax (STX),Syntax (STX),Functionality (FNC),Functionality (FNC),Synthesis (SYN),Synthesis (SYN),Power,Power,Performance,Performance,Area,Area
2
  ,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,VerilogEval S2R,RTLLM,RTL-Repo,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen,VerilogEval MC,VeriGen
3
+ DeepSeek V3.1 Terminus,95.769999999999996,93.620000000000005,75.510000000000005,62.979999999999997,75.510000000000005,58.719999999999999,75.510000000000005,64.200000000000003,72.739999999999995,59.109999999999999,74.420000000000002,62.350000000000001,-1,95.129999999999995,96.469999999999999,77.049999999999997,78.819999999999993,76.790000000000006,75.290000000000006,76.230000000000004,69.870000000000005,74.299999999999997,75.140000000000001,75.75,77.590000000000003
4
+ Kimi K2 Instruct 0905,96.280000000000001,90.209999999999994,71.280000000000001,64.680000000000007,70.769999999999996,60.43,70.069999999999993,62.939999999999998,68.769999999999996,58.350000000000001,70.319999999999993,63.079999999999998,33.649999999999999,98.590000000000003,95.290000000000006,74.739999999999995,69.409999999999997,74.099999999999994,69.409999999999997,73.290000000000006,57.469999999999999,71.099999999999994,69.900000000000006,73,69.510000000000005
5
  Gemini 2.5 Flash (Medium),91.540000000000006,81.700000000000003,69.739999999999995,64.680000000000007,69.230000000000004,46.380000000000003,68.400000000000006,51.109999999999999,66.569999999999993,45.100000000000001,68.269999999999996,49.25,-1,90.640000000000001,96.469999999999999,71.670000000000002,77.650000000000006,71.280000000000001,77.650000000000006,70.280000000000001,71.269999999999996,68.060000000000002,77.349999999999994,70.319999999999993,79.450000000000003
6
  DeepSeek R1-0528,96.540000000000006,89.790000000000006,79.099999999999994,67.659999999999997,78.969999999999999,64.260000000000005,78.439999999999998,73.319999999999993,76.290000000000006,65.379999999999995,78.200000000000003,70.859999999999999,-1,94.739999999999995,98.819999999999993,80,80,79.870000000000005,74.120000000000005,79.659999999999997,63.859999999999999,77.280000000000001,74.049999999999997,79.409999999999997,77.010000000000005
7
  DeepSeek R1,97.689999999999998,93.189999999999998,79.359999999999999,67.659999999999997,79.099999999999994,62.979999999999997,79.010000000000005,71.379999999999995,76.489999999999995,63.740000000000002,78.629999999999995,69.819999999999993,-1,97.439999999999998,95.290000000000006,77.819999999999993,74.120000000000005,77.689999999999998,74.120000000000005,77.299999999999997,64.579999999999998,75.030000000000001,74.420000000000002,76.939999999999998,77.150000000000006
 
67
  ,,,,,,,,,,,,,,,,,,,,,,,,,
68
  ,,,,,,,,,,,,,,,,,,,,,,,,,
69
  ,,,,,,,,,,,,,,,,,,,,,,,,,
 
70
 
static/html_content.py CHANGED
@@ -44,6 +44,9 @@ INTRO_HTML = """
44
  Use the filters below to explore different RTL benchmarks, simulators and models.
45
  </p>
46
  <p style="margin-top:10px; text-align:start !important;">
 
 
 
47
  <span style="font-variant:small-caps; font-weight:bold;">UPDATE (OCT 2025):</span> Added <span>Hermes-4-14B</span>, <span>Qwen3-8B</span>, and <span>Seed-OSS-36B</span> to the leaderboard. Implemented Other Models tab and moved models to it
48
  </p>
49
  <p style="margin-top:-6px; text-align:start !important;">
@@ -115,4 +118,4 @@ OTHER_MODELS_HTML = """
115
  These models were previously listed on the main leaderboard, evaluated with a potentially deprecated version of TuRTLe, and will no longer be updated.
116
  </p>
117
  </div>
118
- """
 
44
  Use the filters below to explore different RTL benchmarks, simulators and models.
45
  </p>
46
  <p style="margin-top:10px; text-align:start !important;">
47
+ <span style="font-variant:small-caps; font-weight:bold;">UPDATE (NOV 2025):</span> We release a new codebase <a href="https://github.com/HPAI-BSC/TuRTLe" target="_blank">TuRTLe v2</a> with API support and local Docker evaluation. Added <span>Kimi K2 Inst</span>, <span>DeepSeek V3.1 Terminus</span>, and <span>Google's Gemini 2.5 Flash</span>
48
+     </p>
49
+ <p style="margin-top:-6px; text-align:start !important;">
50
  <span style="font-variant:small-caps; font-weight:bold;">UPDATE (OCT 2025):</span> Added <span>Hermes-4-14B</span>, <span>Qwen3-8B</span>, and <span>Seed-OSS-36B</span> to the leaderboard. Implemented Other Models tab and moved models to it
51
  </p>
52
  <p style="margin-top:-6px; text-align:start !important;">
 
118
  These models were previously listed on the main leaderboard, evaluated with a potentially deprecated version of TuRTLe, and will no longer be updated.
119
  </p>
120
  </div>
121
+ """
utils.py CHANGED
@@ -15,7 +15,7 @@ def model_hyperlink(link, model_name, release, thinking=False):
15
  new_badge = f' <span class="badge new-badge">new</span>'
16
  reasoning_badge = f' <span class="badge reasoning-badge">reasoning</span>'
17
 
18
- if release == "V3":
19
  # show new badge only to the latest releases
20
  return ret + reasoning_badge + new_badge if thinking == "Reasoning" else ret + new_badge
21
  else:
 
15
  new_badge = f' <span class="badge new-badge">new</span>'
16
  reasoning_badge = f' <span class="badge reasoning-badge">reasoning</span>'
17
 
18
+ if release == "V4":
19
  # show new badge only to the latest releases
20
  return ret + reasoning_badge + new_badge if thinking == "Reasoning" else ret + new_badge
21
  else: