Transformers documentation

MiniMax-M2

You are viewing main version, which requires installation from source. If you'd like regular pip install, checkout the latest stable version (v5.0.0rc2).
Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

MiniMax-M2

Overview

MiniMax-M2 is a compact, fast, and cost-effective MoE model (230 billion total parameters with 10 billion active parameters) built for elite performance in coding and agentic tasks, all while maintaining powerful general intelligence. With just 10 billion activated parameters, MiniMax-M2 provides the sophisticated, end-to-end tool use performance expected from today’s leading models, but in a streamlined form factor that makes deployment and scaling easier than ever.

For more details refer to the release blog post.

Usage examples

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "MiniMaxAI/MiniMax-M2",
    device_map="auto",
    revision="refs/pr/52",
)

tokenizer = AutoTokenizer.from_pretrained("MiniMaxAI/MiniMax-M2", revision="refs/pr/52")

messages = [
    {"role": "user", "content": "What is your favourite condiment?"},
    {"role": "assistant", "content": "Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!"},
    {"role": "user", "content": "Do you have mayonnaise recipes?"}
]

model_inputs = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True).to("cuda")

generated_ids = model.generate(**model_inputs, max_new_tokens=100)

response = tokenizer.batch_decode(generated_ids)[0]

print(response)

MiniMaxM2Config

class transformers.MiniMaxM2Config

< >

( vocab_size: int | None = 200064 hidden_size: int | None = 3072 intermediate_size: int | None = 1536 num_hidden_layers: int | None = 62 num_attention_heads: int | None = 48 num_key_value_heads: int | None = 8 head_dim: int | None = 128 hidden_act: str | None = 'silu' max_position_embeddings: int | None = 196608 initializer_range: float | None = 0.02 rms_norm_eps: int | None = 1e-06 use_cache: bool | None = True pad_token_id: int | None = None bos_token_id: int | None = 200034 eos_token_id: int | None = 200020 tie_word_embeddings: bool | None = False attention_dropout: float | None = 0.0 num_experts_per_tok: int | None = 8 num_local_experts: int | None = 256 output_router_logits: bool | None = False router_aux_loss_coef: float | None = 0.001 router_jitter_noise: float | None = 0.0 rope_parameters: transformers.modeling_rope_utils.RopeParameters | dict[transformers.modeling_rope_utils.RopeParameters] | None = None **kwargs )

Parameters

  • vocab_size (Optional, optional, defaults to 200064) — Vocabulary size of the MiniMaxM2 model. Defines the number of different tokens that can be represented by the inputs_ids passed when calling MiniMaxM2Model
  • hidden_size (Optional, optional, defaults to 3072) — Dimension of the hidden representations.
  • intermediate_size (Optional, optional, defaults to 1536) — Dimension of the MLP representations.
  • num_hidden_layers (Optional, optional, defaults to 62) — Number of hidden layers in the Transformer encoder.
  • num_attention_heads (Optional, optional, defaults to 48) — Number of attention heads for each attention layer in the Transformer encoder.
  • num_key_value_heads (Optional, optional, defaults to 8) — This is the number of key_value heads that should be used to implement Grouped Query Attention. If num_key_value_heads=num_attention_heads, the model will use Multi Head Attention (MHA), if num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group. For more details, check out this paper. If it is not specified, will default to 8.
  • head_dim (Optional, optional, defaults to 128) — The attention head dimension.
  • hidden_act (str or function, optional, defaults to "silu") — The non-linear activation function (function or string) in the decoder.
  • max_position_embeddings (Optional, optional, defaults to 196608) — The maximum sequence length that this model might ever be used with. MiniMaxM2’s sliding window attention allows sequence of up to 196608 tokens.
  • initializer_range (Optional, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
  • rms_norm_eps (Optional, optional, defaults to 1e-06) — The epsilon used by the rms normalization layers.
  • use_cache (Optional, optional, defaults to True) — Whether or not the model should return the last key/values attentions (not used by all models). Only relevant if config.is_decoder=True.
  • pad_token_id (Optional, optional) — The id of the padding token.
  • bos_token_id (Optional, optional, defaults to 200034) — The id of the “beginning-of-sequence” token.
  • eos_token_id (Optional, optional, defaults to 200020) — The id of the “end-of-sequence” token.
  • tie_word_embeddings (bool, optional, defaults to False) — Whether the model’s input and output word embeddings should be tied.
  • attention_dropout (Optional, optional, defaults to 0.0) — The dropout ratio for the attention probabilities.
  • num_experts_per_tok (Optional, optional, defaults to 8) — The number of experts to route per-token, can be also interpreted as the top-k routing parameter
  • num_local_experts (Optional, optional, defaults to 256) — Number of experts per Sparse MLP layer.
  • output_router_logits (Optional, optional, defaults to False) — Whether or not the router logits should be returned by the model. Enabling this will also allow the model to output the auxiliary loss. See here for more details
  • router_aux_loss_coef (Optional, optional, defaults to 0.001) — The aux loss factor for the total loss.
  • router_jitter_noise (Optional, optional, defaults to 0.0) — Amount of noise to add to the router.
  • rope_parameters (RopeParameters, optional) — Dictionary containing the configuration parameters for the RoPE embeddings. The dictionaty should contain a value for rope_theta and optionally parameters used for scaling in case you want to use RoPE with longer max_position_embeddings.

This is the configuration class to store the configuration of a MiniMaxM2Model. It is used to instantiate an MiniMaxM2 model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the MiniMaxM2.

MiniMaxAI/MiniMax-M2

Configuration objects inherit from PreTrainedConfig and can be used to control the model outputs. Read the documentation from PreTrainedConfig for more information.

>>> from transformers import MiniMaxM2Model, MiniMaxM2Config

>>> # Initializing a MiniMaxM2 style configuration
>>> configuration = MiniMaxM2Config()

>>> # Initializing a model from the MiniMaxM2 style configuration
>>> model = MiniMaxM2Model(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

MiniMaxM2Model

class transformers.MiniMaxM2Model

< >

( config: MiniMaxM2Config )

Parameters

  • config (MiniMaxM2Config) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

The bare Minimax M2 Model outputting raw hidden-states without any specific head on top.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

forward

< >

( input_ids: torch.LongTensor | None = None attention_mask: torch.Tensor | None = None position_ids: torch.LongTensor | None = None past_key_values: transformers.cache_utils.Cache | None = None inputs_embeds: torch.FloatTensor | None = None use_cache: bool | None = None cache_position: torch.LongTensor | None = None **kwargs: typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs] ) transformers.modeling_outputs.MoeModelOutputWithPast or tuple(torch.FloatTensor)

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

    Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

    What are input IDs?

  • attention_mask (torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

    • 1 for tokens that are not masked,
    • 0 for tokens that are masked.

    What are attention masks?

  • position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.n_positions - 1].

    What are position IDs?

  • past_key_values (~cache_utils.Cache, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in the past_key_values returned by the model at a previous stage of decoding, when use_cache=True or config.use_cache=True.

    Only Cache instance is allowed as input, see our kv cache guide. If no past_key_values are passed, DynamicCache will be initialized by default.

    The model will output the same cache format that is fed as input.

    If past_key_values are used, the user is expected to input only unprocessed input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, unprocessed_length) instead of all input_ids of shape (batch_size, sequence_length).

  • inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
  • use_cache (bool, optional) — If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values).
  • cache_position (torch.LongTensor of shape (sequence_length), optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily to position_ids, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length.

Returns

transformers.modeling_outputs.MoeModelOutputWithPast or tuple(torch.FloatTensor)

A transformers.modeling_outputs.MoeModelOutputWithPast or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (MiniMaxM2Config) and inputs.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • past_key_values (Cache, optional, returned when use_cache=True is passed or when config.use_cache=True) — It is a Cache instance. For more details, see our kv cache guide.

    Contains pre-computed hidden-states (key and values in the self-attention blocks and optionally if config.is_encoder_decoder=True in the cross-attention blocks) that can be used (see past_key_values input) to speed up sequential decoding.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

  • router_logits (tuple(torch.FloatTensor), optional, returned when output_router_probs=True and config.add_router_probs=True is passed or when config.output_router_probs=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, sequence_length, num_experts).

    Raw router logtis (post-softmax) that are computed by MoE routers, these terms are used to compute the auxiliary loss for Mixture of Experts models.

The MiniMaxM2Model forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

MiniMaxM2ForCausalLM

class transformers.MiniMaxM2ForCausalLM

< >

( config )

Parameters

  • config (MiniMaxM2ForCausalLM) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

The Minimax M2 Model for causal language modeling.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

forward

< >

( input_ids: torch.LongTensor | None = None attention_mask: torch.Tensor | None = None position_ids: torch.LongTensor | None = None past_key_values: transformers.cache_utils.Cache | None = None inputs_embeds: torch.FloatTensor | None = None labels: torch.LongTensor | None = None use_cache: bool | None = None output_router_logits: bool | None = None cache_position: torch.LongTensor | None = None logits_to_keep: int | torch.Tensor = 0 **kwargs: typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs] ) transformers.modeling_outputs.MoeCausalLMOutputWithPast or tuple(torch.FloatTensor)

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

    Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

    What are input IDs?

  • attention_mask (torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

    • 1 for tokens that are not masked,
    • 0 for tokens that are masked.

    What are attention masks?

  • position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.n_positions - 1].

    What are position IDs?

  • past_key_values (~cache_utils.Cache, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in the past_key_values returned by the model at a previous stage of decoding, when use_cache=True or config.use_cache=True.

    Only Cache instance is allowed as input, see our kv cache guide. If no past_key_values are passed, DynamicCache will be initialized by default.

    The model will output the same cache format that is fed as input.

    If past_key_values are used, the user is expected to input only unprocessed input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, unprocessed_length) instead of all input_ids of shape (batch_size, sequence_length).

  • inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
  • labels (torch.LongTensor of shape (batch_size, sequence_length), optional) — Labels for computing the masked language modeling loss. Indices should either be in [0, ..., config.vocab_size] or -100 (see input_ids docstring). Tokens with indices set to -100 are ignored (masked), the loss is only computed for the tokens with labels in [0, ..., config.vocab_size].
  • use_cache (bool, optional) — If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values).
  • output_router_logits (bool, optional) — Whether or not to return the logits of all the routers. They are useful for computing the router loss, and should not be returned during inference.
  • cache_position (torch.LongTensor of shape (sequence_length), optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily to position_ids, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length.
  • logits_to_keep (Union[int, torch.Tensor], optional, defaults to 0) — If an int, compute logits for the last logits_to_keep tokens. If 0, calculate logits for all input_ids (special case). Only last token logits are needed for generation, and calculating them only for that token can save memory, which becomes pretty significant for long sequences or large vocabulary size. If a torch.Tensor, must be 1D corresponding to the indices to keep in the sequence length dimension. This is useful when using packed tensor format (single dimension for batch and sequence length).

Returns

transformers.modeling_outputs.MoeCausalLMOutputWithPast or tuple(torch.FloatTensor)

A transformers.modeling_outputs.MoeCausalLMOutputWithPast or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (MiniMaxM2Config) and inputs.

  • loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Language modeling loss (for next-token prediction).

  • logits (torch.FloatTensor of shape (batch_size, sequence_length, config.vocab_size)) — Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).

  • aux_loss (torch.FloatTensor, optional, returned when labels is provided) — aux_loss for the sparse modules.

  • router_logits (tuple(torch.FloatTensor), optional, returned when output_router_probs=True and config.add_router_probs=True is passed or when config.output_router_probs=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, sequence_length, num_experts).

    Raw router logtis (post-softmax) that are computed by MoE routers, these terms are used to compute the auxiliary loss for Mixture of Experts models.

  • past_key_values (Cache, optional, returned when use_cache=True is passed or when config.use_cache=True) — It is a Cache instance. For more details, see our kv cache guide.

    Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see past_key_values input) to speed up sequential decoding.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

The MiniMaxM2ForCausalLM forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example:

>>> from transformers import AutoTokenizer, MiniMaxM2ForCausalLM

>>> model = MiniMaxM2ForCausalLM.from_pretrained("mistralai/MiniMaxM2-8x7B-v0.1")
>>> tokenizer = AutoTokenizer.from_pretrained("mistralai/MiniMaxM2-8x7B-v0.1")

>>> prompt = "Hey, are you conscious? Can you talk to me?"
>>> inputs = tokenizer(prompt, return_tensors="pt")

>>> # Generate
>>> generate_ids = model.generate(inputs.input_ids, max_length=30)
>>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
"Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
Update on GitHub