Hugging Face's logo

Transformers documentation

BitNet

You are viewing main version, which requires installation from source. If you'd like regular pip install, checkout the latest stable version (v4.51.3).
Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

BitNet

Overview

Trained on a corpus of 4 trillion tokens, this model demonstrates that native 1-bit LLMs can achieve performance comparable to leading open-weight, full-precision models of similar size, while offering substantial advantages in computational efficiency (memory, energy, latency).

➡️ Technical Report: BitNet b1.58 2B4T Technical Report

➡️ Official Inference Code: microsoft/BitNet (bitnet.cpp)

Model Variants

Several versions of the model weights are available on Hugging Face:

Model Details

  • Architecture: Transformer-based, modified with BitLinear layers (BitNet framework).
    • Uses Rotary Position Embeddings (RoPE).
    • Uses squared ReLU (ReLU²) activation in FFN layers.
    • Employs subln normalization.
    • No bias terms in linear or normalization layers.
  • Quantization: Native 1.58-bit weights and 8-bit activations (W1.58A8).
    • Weights are quantized to ternary values {-1, 0, +1} using absmean quantization during the forward pass.
    • Activations are quantized to 8-bit integers using absmax quantization (per-token).
    • Crucially, the model was trained from scratch with this quantization scheme, not post-training quantized.
  • Parameters: ~2 Billion
  • Training Tokens: 4 Trillion
  • Context Length: Maximum sequence length of 4096 tokens.
    • Recommendation: For optimal performance on tasks requiring very long contexts (beyond the pre-training length or for specialized long-reasoning tasks), we recommend performing intermediate long-sequence adaptation/training before the final fine-tuning stage.
  • Training Stages:
    1. Pre-training: Large-scale training on public text/code and synthetic math data using a two-stage learning rate and weight decay schedule.
    2. Supervised Fine-tuning (SFT): Fine-tuned on instruction-following and conversational datasets using sum loss aggregation and specific hyperparameter tuning.
    3. Direct Preference Optimization (DPO): Aligned with human preferences using preference pairs.
  • Tokenizer: LLaMA 3 Tokenizer (vocab size: 128,256).

Usage tips

VERY IMPORTANT NOTE ON EFFICIENCY

Please do NOT expect performance efficiency gains (in terms of speed, latency, or energy consumption) when using this model with the standard transformers library.

The current execution paths within transformers do not contain the specialized, highly optimized computational kernels required to leverage the advantages of the BitNet architecture. Running the model via transformers will likely result in inference speeds and energy usage comparable to, or potentially worse than, standard full-precision models within this framework on both CPU and GPU.

While you might observe reduced memory usage due to the quantized weights, the primary computational efficiency benefits are not accessible through this standard transformers usage path.

For achieving the efficiency benefits demonstrated in the technical paper, you MUST use the dedicated C++ implementation: bitnet.cpp.

Requirements

pip install transformers

Example

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "microsoft/bitnet-b1.58-2B-4T"

# Load tokenizer and model
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16
)

# Apply the chat template
messages = [
    {"role": "system", "content": "You are a helpful AI assistant."},
    {"role": "user", "content": "How are you?"},
]
chat_input = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to(model.device)

# Generate response
chat_outputs = model.generate(chat_input, max_new_tokens=50)
response = tokenizer.decode(chat_outputs[0][chat_input.shape[-1]:], skip_special_tokens=True) # Decode only the response part
print("\nAssistant Response:", response)

BitNetConfig

class transformers.BitNetConfig

< >

( vocab_size = 128256 hidden_size = 2560 intermediate_size = 6912 num_hidden_layers = 30 num_attention_heads = 20 num_key_value_heads = 5 hidden_act = 'relu2' max_position_embeddings = 2048 initializer_range = 0.02 rms_norm_eps = 1e-05 use_cache = True pad_token_id = None bos_token_id = 128000 eos_token_id = 128001 tie_word_embeddings = False rope_theta = 500000.0 attention_bias = False attention_dropout = 0.0 **kwargs )

Parameters

  • vocab_size (int, optional, defaults to 128256) — Vocabulary size of the BitNet model. Defines the number of different tokens that can be represented by the inputs_ids passed when calling BitNetModel
  • hidden_size (int, optional, defaults to 2560) — Dimension of the hidden representations.
  • intermediate_size (int, optional, defaults to 6912) — Dimension of the MLP representations.
  • num_hidden_layers (int, optional, defaults to 30) — Number of hidden layers in the Transformer decoder.
  • num_attention_heads (int, optional, defaults to 20) — Number of attention heads for each attention layer in the Transformer decoder.
  • num_key_value_heads (int, optional, defaults to 5) — 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 checkout [this paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to num_attention_heads`.
  • hidden_act (str or function, optional, defaults to "relu2") — The non-linear activation function (function or string) in the decoder.
  • max_position_embeddings (int, optional, defaults to 2048) — The maximum sequence length that this model might ever be used with.
  • initializer_range (float, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
  • rms_norm_eps (float, optional, defaults to 1e-05) — The epsilon used by the rms normalization layers.
  • use_cache (bool, 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 (int, optional) — Padding token id.
  • bos_token_id (int, optional, defaults to 128000) — Beginning of stream token id.
  • eos_token_id (int, optional, defaults to 128001) — End of stream token id.
  • tie_word_embeddings (bool, optional, defaults to False) — Whether to tie weight embeddings
  • rope_theta (float, optional, defaults to 500000.0) — The base period of the RoPE embeddings.
  • attention_bias (bool, optional, defaults to False) — Whether to use a bias in the query, key, value and output projection layers during self-attention.
  • attention_dropout (float, optional, defaults to 0.0) — The dropout ratio for the attention probabilities.

This is the configuration class to store the configuration of a BitNetModel. It is used to instantiate an BitNet model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of BitNet b1.58 2B4T microsoft/bitnet-b1.58-2B-4T.

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 BitNetModel, BitNetConfig

>>> # Initializing a BitNet style configuration
>>> configuration = BitNetConfig()

>>> # Initializing a model from the BitNet style configuration
>>> model = BitNetModel(configuration)

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

BitNetModel

class transformers.BitNetModel

< >

( config: BitNetConfig )

Parameters

  • config (BitNetConfig) — 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.
  • config — BitNetConfig

The bare BitNet 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.

Transformer decoder consisting of config.num_hidden_layers layers. Each layer is a BitNetDecoderLayer

forward

< >

( input_ids: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Optional[transformers.cache_utils.Cache] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None cache_position: typing.Optional[torch.LongTensor] = None **flash_attn_kwargs: typing_extensions.Unpack[transformers.modeling_flash_attention_utils.FlashAttentionKwargs] )

Parameters

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

    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) or BlockMask, *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.

    If the model is configured to use flex_attention, it will attempt to convert the mask Tensor into a BlockMask, but you can also pass a BlockMask object directly here.

    What are attention masks?

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

    If past_key_values is used, optionally only the last input_ids have to be input (see past_key_values).

    If you want to change padding behavior, you should read modeling_opt._prepare_decoder_attention_mask and modify to your needs. See diagram 1 in the paper for more information on the default strategy.

    • 1 indicates the head is not masked,
    • 0 indicates the head is masked.
  • 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, 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.

    It is a Cache instance. For more details, see our kv cache guide.

    If past_key_values are used, the user can optionally input only the last input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, 1) 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).
  • output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
  • output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
  • return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
  • 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.

The BitNetModel 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.

BitNetForCausalLM

class transformers.BitNetForCausalLM

< >

( config )

forward

< >

( input_ids: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Optional[transformers.cache_utils.Cache] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None labels: typing.Optional[torch.LongTensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None cache_position: typing.Optional[torch.LongTensor] = None logits_to_keep: typing.Union[int, torch.Tensor] = 0 **kwargs: typing_extensions.Unpack[transformers.models.bitnet.modeling_bitnet.KwargsForCausalLM] ) → transformers.modeling_outputs.CausalLMOutputWithPast or tuple(torch.FloatTensor)

Parameters

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

    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) or BlockMask, *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.

    If the model is configured to use flex_attention, it will attempt to convert the mask Tensor into a BlockMask, but you can also pass a BlockMask object directly here.

    What are attention masks?

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

    If past_key_values is used, optionally only the last input_ids have to be input (see past_key_values).

    If you want to change padding behavior, you should read modeling_opt._prepare_decoder_attention_mask and modify to your needs. See diagram 1 in the paper for more information on the default strategy.

    • 1 indicates the head is not masked,
    • 0 indicates the head is masked.
  • 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, 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.

    It is a Cache instance. For more details, see our kv cache guide.

    If past_key_values are used, the user can optionally input only the last input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, 1) 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).
  • output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
  • output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
  • return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
  • 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.
  • Args — labels (torch.LongTensor of shape (batch_size, sequence_length), optional): Labels for computing the masked language modeling loss. Indices should either be in [0, transformers., 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, transformers., config.vocab_size].

Returns

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

A transformers.modeling_outputs.CausalLMOutputWithPast 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 (BitNetConfig) 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).

  • past_key_values (tuple(tuple(torch.FloatTensor)), optional, returned when use_cache=True is passed or when config.use_cache=True) — Tuple of tuple(torch.FloatTensor) of length config.n_layers, with each tuple having 2 tensors of shape (batch_size, num_heads, sequence_length, embed_size_per_head))

    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 BitNetForCausalLM 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, BitNetForCausalLM

>>> model = BitNetForCausalLM.from_pretrained("microsoft/bitnet-b1.58-2B-4T")
>>> tokenizer = AutoTokenizer.from_pretrained("microsoft/bitnet-b1.58-2B-4T")

>>> prompt = f'<|begin_of_text|>User: Hey, are you conscious? Can you talk to me?<|eot_id|>Assistant: '
>>> inputs = tokenizer(prompt, return_tensors="pt")

>>> # Generate
>>> generate_ids = model.generate(inputs.input_ids, max_length=100)
>>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
"User: Hey, are you conscious? Can you talk to me?Assistant: No, I'm not conscious. I'm an artificial intelligence designed to assist with information and tasks. How can I help you today?"
< > Update on GitHub

←BioGpt Blenderbot→