Transformers v5 release notes

• Highlights
• Significant API changes: dynamic weight loading, tokenization
• Backwards Incompatible Changes
• Bugfixes and improvements

We have a migration guide that will be continuously updated available on the main branch, please check it out in case you're facing issues: migration guide.

Highlights

We are excited to announce the initial release of Transformers v5. This is the first major release in five years, and the release is significant: 1200 commits have been pushed to main since the latest minor release. This release removes a lot of long-due deprecations, introduces several refactors that significantly simplify our APIs and internals, and comes with a large number of bug fixes.

We give an overview of our focus for this release in the following blogpost. In these release notes, we'll focus directly on the refactors and new APIs coming with v5.

This release is the full V5 release. It sets in motion something bigger: going forward, starting with v5, we'll now release minor releases every week, rather than every 5 weeks. Expect v5.1 to follow next week, then v5.2 the week that follows, etc.

We're moving forward with this change to ensure you have access to models as soon as they're supported in the library, rather than a few weeks after.

In order to install this release, please do so with the following:

pip install transformers

For us to deliver the best package possible, it is imperative that we have feedback on how the toolkit is currently working for you. Please try it out, and open an issue in case you're facing something inconsistent/a bug.

Transformers version 5 is a community endeavor, and we couldn't have shipped such a massive release without the help of the entire community.

Significant API changes

Dynamic weight loading

We introduce a new weight loading API in transformers, which significantly improves on the previous API. This weight loading API is designed to apply operations to the checkpoints loaded by transformers.

Instead of loading the checkpoint exactly as it is serialized within the model, these operations can reshape, merge, and split the layers according to how they're defined in this new API. These operations are often a necessity when working with quantization or parallelism algorithms.

This new API is centered around the new WeightConverter class:

class WeightConverter(WeightTransform):
    operations: list[ConversionOps]
    source_keys: Union[str, list[str]]
    target_keys: Union[str, list[str]]

The weight converter is designed to apply a list of operations on the source keys, resulting in target keys. A common operation done on the attention layers is to fuse the query, key, values layers. Doing so with this API would amount to defining the following conversion:

conversion = WeightConverter(
    ["self_attn.q_proj", "self_attn.k_proj", "self_attn.v_proj"],  # The input layers
    "self_attn.qkv_proj",  # The single layer as output
    operations=[Concatenate(dim=0)],
)

In this situation, we apply the Concatenate operation, which accepts a list of layers as input and returns a single layer.

This allows us to define a mapping from architecture to a list of weight conversions. Applying those weight conversions can apply arbitrary transformations to the layers themselves. This significantly simplified the from_pretrained method and helped us remove a lot of technical debt that we accumulated over the past few years.

This results in several improvements:

• Much cleaner definition of transformations applied to the checkpoint
• Reversible transformations, so loading and saving a checkpoint should result in the same checkpoint
• Faster model loading thanks to scheduling of tensor materialization
• Enables complex mix of transformations that wouldn't otherwise be possible (such as quantization + MoEs, or TP + MoEs)

Linked PR: https://github.com/huggingface/transformers/pull/41580

Tokenization

Just as we moved towards a single backend library for model definition, we want our tokenizers, and the Tokenizer object to be a lot more intuitive. With v5, tokenizer definition is much simpler; one can now initialize an empty LlamaTokenizer and train it directly on your corpus.

Defining a new tokenizer object should be as simple as this:

from transformers import TokenizersBackend, generate_merges
from tokenizers import pre_tokenizers, Tokenizer
from tokenizers.model import BPE

class Llama5Tokenizer(TokenizersBackend):
    def __init__(self, unk_token="<unk>",bos_token="<s>", eos_token="</s>", vocab=None, merges=None ):
        if vocab is None:
            self._vocab = {
                str(unk_token): 0,
                str(bos_token): 1,
                str(eos_token): 2,
            }

        else:
            self._vocab = vocab

            self._merges = merges

        self._tokenizer = Tokenizer(
            BPE(vocab=self._vocab, merges=self._merges, fuse_unk=True)
        )
        self._tokenizer.pre_tokenizer = pre_tokenizers.Metaspace(
            replacement="▁", prepend_scheme=_get_prepend_scheme(self.add_prefix_space, self), split=False
        )
        super().__init__(
            tokenizer_object=self._tokenizer,
            unk_token=unk_token,
            bos_token=bos_token,
            eos_token=eos_token,
        )

Once the tokenizer is defined as above, you can load it with the following: Llama5Tokenizer(). Doing this returns you an empty, trainable tokenizer that follows the definition of the authors of Llama5 (it does not exist yet :wink:).

The above is the main motivation towards refactoring tokenization: we want tokenizers to behave similarly to models: trained or empty, and with exactly what is defined in their class definition.

Backend Architecture Changes: moving away from the slow/fast tokenizer separation

Up to now, transformers maintained two parallel implementations for many tokenizers:

• "Slow" tokenizers (tokenization_<model>.py) - Python-based implementations, often using SentencePiece as the backend.
• "Fast" tokenizers (tokenization_<model>_fast.py) - Rust-based implementations using the 🤗 tokenizers library.

In v5, we consolidate to a single tokenizer file per model: tokenization_<model>.py. This file will use the most appropriate backend available:

1. TokenizersBackend (preferred): Rust-based tokenizers from the 🤗 tokenizers library. In general it provides optimal performance, but it also offers a lot more features that are commonly adopted across the ecosystem:

• handling additional tokens
• a full python API for setting and updating
• automatic parallelization,
• automatic offsets
• customization
• training

2. SentencePieceBackend: for tokenizers requiring the sentencepiece library. It inherits from PythonBackend.
3. PythonBackend: a Python implementations of the features provided by tokenizers. Basically allows adding tokens.
4. MistralCommonBackend: relies on MistralCommon's tokenization library. (Previously known as the MistralCommonTokenizer)

The AutoTokenizer automatically selects the appropriate backend based on available files and dependencies. This is transparent, you continue to use AutoTokenizer.from_pretrained() as before. This allows transformers to be future-proof and modular to easily support future backends.

Defining a tokenizers outside of the existing backends

We enable users and tokenizer builders to define their own tokenizers from top to bottom. Tokenizers are usually defined using a backend such as tokenizers, sentencepiece or mistral-common, but we offer the possibility to design the tokenizer at a higher-level, without relying on those backends.

To do so, you can import the PythonBackend (which was previously known as PreTrainedTokenizer). This class encapsulates all the logic related to added tokens, encoding, and decoding.

If you want something even higher up the stack, then PreTrainedTokenizerBase is what PythonBackend inherits from. It contains the very basic tokenizer API features:

• encode
• decode
• vocab_size
• get_vocab
• convert_tokens_to_ids
• convert_ids_to_tokens
• from_pretrained
• save_pretrained
• among a few others

API Changes

1. Direct tokenizer initialization with vocab and merges

Starting with v5, we now enable initializing blank, untrained tokenizers-backed tokenizers:

from transformers import LlamaTokenizer

tokenizer = LlamaTokenizer()

This tokenizer will therefore follow the definition of the LlamaTokenizer as defined in its class definition. It can then be trained on a corpus as can be seen in the tokenizers documentation.

These tokenizers can also be initialized from vocab and merges (if necessary), like the previous "slow" tokenizers:

from transformers import LlamaTokenizer

vocab = {"<unk>": 0, "<s>": 1, "</s>": 2, "hello": 3, "world": 4}
merges = [("h", "e"), ("l", "l"), ("o", " ")]

tokenizer = LlamaTokenizer(vocab=vocab, merges=merges)

This tokenizer will behave as a Llama-like tokenizer, with an updated vocabulary. This allows comparing different tokenizer classes with the same vocab; therefore enabling the comparison of different pre-tokenizers, normalizers, etc.

⚠️ The vocab_file (as in, a path towards a file containing the vocabulary) cannot be used to initialize the LlamaTokenizer as loading from files is reserved to the from_pretrained method.

2. Simplified decoding API

The batch_decode and decode methods have been unified to reflect behavior of the encode method. Both single and batch decoding now use the same decode method. See an example of the new behavior below:

from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("t5-small") 
inputs = ["hey how are you?", "fine"]
tokenizer.decode(tokenizer.encode(inputs))

Gives:

- 'hey how are you?</s> fine</s>'
+ ['hey how are you?</s>', 'fine</s>']

We expect encode and decode to behave, as two sides of the same coin: encode, process, decode, should work.

[!NOTE] A common use-case would be: encode, model.generate, decode. However, using generate would return list[list[int]], which would then be incompatible with decode.

3. Unified encoding API

The encode_plus method is deprecated in favor of the single __call__ method.

4. apply_chat_template returns BatchEncoding

Previously, apply_chat_template returned input_ids for backward compatibility. Starting with v5, it now consistently returns a BatchEncoding dict like other tokenizer methods.

# v5
messages = [
    {"role": "user", "content": "Hello!"},
    {"role": "assistant", "content": "Hi there!"}
]

# Now returns BatchEncoding with input_ids, attention_mask, etc.
outputs = tokenizer.apply_chat_template(messages, return_tensors="pt")
print(outputs.keys())  # dict_keys(['input_ids', 'attention_mask'])

5. Removed legacy configuration file saving:

We simplify the serialization of tokenization attributes:

• special_tokens_map.json - special tokens are now stored in tokenizer_config.json.
• added_tokens.json - added tokens are now stored in tokenizer.json.
• added_tokens_decoder is only stored when there is no tokenizer.json.

When loading older tokenizers, these files are still read for backward compatibility, but new saves use the consolidated format. We're gradually moving towards consolidating attributes to fewer files so that other libraries and implementations may depend on them more reliably.

6. Model-Specific Changes

Several models that had identical tokenizers now import from their base implementation:

• LayoutLM → uses BertTokenizer
• LED → uses BartTokenizer
• Longformer → uses RobertaTokenizer
• LXMert → uses BertTokenizer
• MT5 → uses T5Tokenizer
• MVP → uses BartTokenizer

These modules will eventually be removed altogether.

Removed T5-specific workarounds

The internal _eventually_correct_t5_max_length method has been removed. T5 tokenizers now handle max length consistently with other models.

Testing Changes

A few testing changes specific to tokenizers have been applied:

• Model-specific tokenization test files now focus on integration tests.
• Common tokenization API tests (e.g., add_tokens, encode, decode) are now centralized and automatically applied across all tokenizers. This reduces test duplication and ensures consistent behavior

For legacy implementations, the original BERT Python tokenizer code (including WhitespaceTokenizer, BasicTokenizer, etc.) is preserved in bert_legacy.py for reference purposes.

7. Deprecated / Modified Features

Special Tokens Structure:

• SpecialTokensMixin: Merged into PreTrainedTokenizerBase to simplify the tokenizer architecture.
• special_tokens_map: Now only stores named special token attributes (e.g., bos_token, eos_token). Use extra_special_tokens for additional special tokens (formerly additional_special_tokens). all_special_tokens includes both named and extra tokens.

# v4
tokenizer.special_tokens_map  # Included 'additional_special_tokens'

# v5
tokenizer.special_tokens_map  # Only named tokens
tokenizer.extra_special_tokens  # Additional tokens

• special_tokens_map_extended and all_special_tokens_extended: Removed. Access AddedToken objects directly from _special_tokens_map or _extra_special_tokens if needed.
• additional_special_tokens: Still accepted for backward compatibility but is automatically converted to extra_special_tokens.

Deprecated Methods:

• sanitize_special_tokens(): Already deprecated in v4, removed in v5.
• prepare_seq2seq_batch(): Deprecated; use __call__() with text_target parameter instead.

# v4
model_inputs = tokenizer.prepare_seq2seq_batch(src_texts, tgt_texts, max_length=128)

# v5
model_inputs = tokenizer(src_texts, text_target=tgt_texts, max_length=128, return_tensors="pt")
model_inputs["labels"] = model_inputs.pop("input_ids_target")

• BatchEncoding.words(): Deprecated; use word_ids() instead.

Removed Methods:

• create_token_type_ids_from_sequences(): Removed from base class. Subclasses that need custom token type ID creation should implement this method directly.
• prepare_for_model(), build_inputs_with_special_tokens(), truncate_sequences(): Moved from tokenization_utils_base.py to tokenization_python.py for PythonBackend tokenizers. TokenizersBackend provides model-ready input via tokenize() and encode(), so these methods are no longer needed in the base class.
• _switch_to_input_mode(), _switch_to_target_mode(), as_target_tokenizer(): Removed from base class. Use __call__() with text_target parameter instead.

# v4
with tokenizer.as_target_tokenizer():
    labels = tokenizer(tgt_texts, ...)

# v5
labels = tokenizer(text_target=tgt_texts, ...)

• parse_response(): Removed from base class.

Performance

MoE Performance

The v5 release significantly improves the performance of the MoE models, as can be seen in the graphs below. We improve and optimize MoE performance through batched and grouped experts implementations, and we optimize them for decoding using batched_mm.

Core performance

We focus on improving the performance of loading weights on device (which gives speedups up to 6x in tensor parallel situations); this is preliminary work that we'll continue to work on in the coming weeks. Some notable improvements:

• [saving] Simplify general logic by @Cyrilvallez in https://github.com/huggingface/transformers/pull/42766
• Do not rely on config for inferring model dtype by @Cyrilvallez in https://github.com/huggingface/transformers/pull/42838
• Improve BatchFeature: stack list and lists of torch tensors by @yonigozlan in https://github.com/huggingface/transformers/pull/42750
• Remove tied weights from internal attribute if they are not tied by @Cyrilvallez in https://github.com/huggingface/transformers/pull/42871
• Enforce call to post_init and fix all of them by @Cyrilvallez in https://github.com/huggingface/transformers/pull/42873
• Simplify tie weights logic by @Cyrilvallez in https://github.com/huggingface/transformers/pull/42895
• Add buffers to _init_weights for ALL models by @Cyrilvallez in https://github.com/huggingface/transformers/pull/42309
• [loading] Really initialize on meta device for huge perf gains by @Cyrilvallez in https://github.com/huggingface/transformers/pull/42941
• Do not use accelerate hooks if the device_map has only 1 device by @Cyrilvallez in https://github.com/huggingface/transformers/pull/43019
• Move missing weights and non-persistent buffers to correct device earlier by @Cyrilvallez in https://github.com/huggingface/transformers/pull/43021

Library-wide changes with lesser impact

Default dtype update

We have updated the default dtype for all models loaded with from_pretrained to be auto. This will lead to model instantiations respecting the dtype in which the model was saved, rather than forcing it to load in float 32.

You can, of course, still specify the dtype in which you want to load your model by specifying it as an argument to the from_pretrained method.

Shard size

The Hugging Face Hub infrastructure has gradually moved to a XET backend. This will significantly simplify uploads and downloads, with higher download and upload speeds, partial uploads, and, most notably, a higher threshold for accepted file sizes on the Hugging Face Hub.

To reflect this, we're increasing the default shard size of models serialized on the Hub to 50GB (up from 5GB).

use_auth_token

The use_auth_token argument/parameter is deprecated in favor of token everywhere. You should be able to search and replace use_auth_token with token and get the same logic.

Linked PR: https://github.com/huggingface/transformers/pull/41666

Attention-related features

We decided to remove some features for the upcoming v5 as they are currently only supported in a few old models and no longer integrated in current model additions. It's recommended to stick to v4.x in case you need them. Following features are affected:

• No more head masking, see #41076. This feature allowed to turn off certain heads during the attention calculation and only worked for eager.
• No more relative positional biases in Bert-like models, see #41170. This feature was introduced to allow relative position scores within attention calculations (similar to T5). However, this feature is barely used in official models and a lot of complexity instead. It also only worked with eager.
• No more head pruning, see #41417 by @gante. As the name suggests, it allowed to prune heads within your attention layers.

Updates to supported torch APIs

We dropped support for two torch APIs:

• torchscript in https://github.com/huggingface/transformers/pull/41688
• torch.fx in https://github.com/huggingface/transformers/pull/41683

Those APIs were deprecated by the PyTorch team, and we're instead focusing on the supported APIs dynamo and export.

Quantization changes

We clean up the quantization API in transformers, and significantly refactor the weight loading as highlighted above.

We drop support for two quantization arguments that have been deprecated for some time:

• load_in_4bit
• load_in_8bit

We remove them in favor of the quantization_config argument which is much more complete. As an example, here is how you would load a 4-bit bitsandbytes model using this argument:

from transformers import AutoModelForCausalLM, BitsAndBytesConfig

quantization_config = BitsAndBytesConfig(load_in_4bit=True)

model_4bit = AutoModelForCausalLM.from_pretrained(
    "meta-llama/Llama-3.2-3B",
    device_map="auto",
    quantization_config=quantization_config
)

Configuration

• Methods to init a nested config such as from_xxx_config are deleted. Configs can be init from the __init__ method in the same way. See #41314.
• It is no longer possible to load a config class from a URL file. Configs must be loaded from either a local path or a repo on the Hub. See #42383.
• All parameters for configuring model's rotary embedding are now stored under mode.rope_parameters, including the rope_theta and rope_type. Model's config.rope_parameters is a simple dictionaty in most cases, and can also be a nested dict in special cases (i.e. Gemma3 and ModernBert) with different rope parameterization for each layer type. Trying to get config.rope_theta will throw an attribute error from now on. See #39847 and #42255
• Qwen-VL family configuration is in a nested format and trying to access keys directly will throw an error (e.g. config.vocab_size). Users are expected to access keys from their respective sub-configs (config.text_config.vocab_size).
• Configurations of non-generative models (any model that doesn't call model.generate()) will no longer have a generation_config and model.config.generation_config will throw an attribute error.

Processing

Tokenization

• Slow tokenizer files (aka: tokenization_<model>.py ) will be removed in favor of using fast tokenizer files tokenization_<model>_fast.py --> will be renamed to tokenization_<model>.py. As fast tokenizers are :hugs:tokenizers - backend, they include a wider range of features that are maintainable and reliable.
• Other backends (sentence piece, tokenizers, etc.) will be supported with a light layer if loading a fast tokenizer fails
• Remove legacy files like special_tokens_map.json and added_tokens.json
• Remove _eventually_correct_t5_max_length
• encode_plus --> __call__
• batch_decode --> decode

apply_chat_template by default returns naked input_ids rather than a BatchEncoding dict. This was inconvenient - it should return a BatchEncoding dict like tokenizer.__call__(), but we were stuck with it for backward compatibility. The method now returns a BatchEncoding.

Linked PRs:

• https://github.com/huggingface/transformers/issues/40938
• https://github.com/huggingface/transformers/pull/40936
• https://github.com/huggingface/transformers/pull/41626

Processing classes

• In processing classes each attribute will be serialized under processor_config.json as a nested dict, instead of serializing attributes in their own config files. Loading will be supported for all old format processors (https://github.com/huggingface/transformers/pull/41474)
• XXXFeatureExtractors classes are completely removed in favor of XXXImageProcessor class for all vision models (https://github.com/huggingface/transformers/pull/41174)
• Minor change: XXXFastImageProcessorKwargs is removed in favor of XXXImageProcessorKwargs which will be shared between fast and slow processors (https://github.com/huggingface/transformers/pull/40931)

Modeling

• Some RotaryEmbeddings layers will start returning a dict of tuples, in case the model uses several RoPE configurations (Gemma2, ModernBert). Each value will be a tuple of "cos, sin" per RoPE type.
• Config attribute for RotaryEmbeddings layer will be unified and accessed via config.rope_parameters. Config attr for rope_theta might not be accessible anymore for some models, and instead will be in config.rope_parameters['rope_theta']. BC will be supported for a while as much as possible, and in the near future we'll gradually move to the new RoPE format (https://github.com/huggingface/transformers/pull/39847)
• Vision Language models will not have a shortcut access to its language and vision component from the generative model via model.language_model. It is recommended to either access the module with model.model.language_model or model.get_decoder(). See #42156
• All models now accept kwargs in their forward methods

Generate

• Old, deprecated output type aliases were removed (e.g. GreedySearchEncoderDecoderOutput). We now only have 4 output classes built from the following matrix: decoder-only vs encoder-decoder, uses beams vs doesn't use beams (https://github.com/huggingface/transformers/pull/40998)
• Removed deprecated classes regarding decoding methods that were moved to the Hub due to low usage (constraints and beam scores) (https://github.com/huggingface/transformers/pull/41223)
• If generate doesn't receive any KV Cache argument, the default cache class used is now defined by the model (as opposed to always being DynamicCache) (https://github.com/huggingface/transformers/pull/41505)
• Generation parameters are no longer accessible via model's config. If generation paramaters are serialized in config.json for any old model, it will be loaded back into model's generation config. Users are expected to access or modify generation parameters only with model.generation_config.do_sample = True.

Trainer

New Features

• ALST/Ulysses Sequence Parallelism Integration
  • Added sequence parallelism support via HF Accelerate for training with longer sequences. Enables splitting sequences across devices using ALST (All-to-All Long Sequence Training) and Ulysses algorithms with DeepSpeed.
• Improved compute_loss_func Handling
  • compute_loss_func now always takes priority over the model's built-in loss computation, giving users consistent control over custom loss functions.
• num_items_in_batch in Prediction Step
  • The num_items_in_batch argument is now passed to compute_loss during prediction_step, enabling proper loss scaling during evaluation.

Breaking Changes

• report_to now defaults to "none"
  • Logging integrations are no longer auto-detected by default; users must explicitly specify which reporting backends to use.

Removing arguments without deprecation cycle in TrainingArguments due to low usage

• mp_parameters -> legacy param that was later on added to the Sagemaker trainer
• _n_gpu -> not intended for users to set, we will initialize it correctly instead of putting it in the TrainingArguments
• overwrite_output_dir - > replaced by resume_from_checkpoint, and it was only used in the examples script, no impact on Trainer.
• logging_dir -> only used for tensorboard, set TENSORBOARD_LOGGING_DIR env var instead
• jit_mode_eval -> use use_torch_compile instead, as torchscript is not recommended anymore
• tpu_num_cores-> It is actually better to remove it, as it is not recommended to set the number of cores. By default, all TPU cores are used . Set TPU_NUM_CORES env var instead
• past_index -> it was only used for a very small number of models that have special architecture like transformersxl + it was not documented at all how to train those models
• ray_scope -> only for a minor arg for ray integration. Set RAY_SCOPE var env instead
• warmup_ratio -> use warmup_step instead. We combined both args together by allowing passing float values in warmup_step.

Removing deprecated arguments in TrainingArguments

• fsdp_min_num_params and fsdp_transformer_layer_cls_to_wrap -> use fsdp_config
• tpu_metrics_debug -> debug
• push_to_hub_token -> hub_token
• push_to_hub_model_id and push_to_hub_organization -> hub_model_id
• include_inputs_for_metrics -> include_for_metrics
• per_gpu_train_batch_size -> per_device_train_batch_size
• per_gpu_eval_batch_size -> per_device_eval_batch_size
• use_mps_device -> mps will be used by default if detected
• fp16_backend and half_precision_backend -> we will only rely on torch.amp as everything has been upstreamed to torch
• no_cuda -> use_cpu
• include_tokens_per_second -> include_num_input_tokens_seen
• use_legacy_prediction_loop -> we only use evaluation_loop function from now on

Removing deprecated arguments in Trainer

• tokenizer in initialization -> processing_class
• model_path in train() -> resume_from_checkpoint

Removed features for Trainer

• sigpot integration for hp search was removed as the library was archived + the api stopped working
• drop support for sagemaker API <1.10
• bump accelerate minimum version to 1.1.0
• bump peft minimum version to 0.18.0
• bump bitsandbytes minimum version to 0.46.1

New defaults for Trainer

• use_cache in the model config will be set to False. You can still change the cache value through TrainingArguments usel_cache argument if needed.

Pipeline

• Image text to text pipelines will no longer accept images as a separate argument along with conversation chats. Image data has to be embedded in the chat's "content" field. See #42359

PushToHubMixin

• removed deprecated organization and repo_url from PushToHubMixin. You must pass a repo_id instead.
• removed ignore_metadata_errors from PushToMixin. In practice if we ignore errors while loading the model card, we won't be able to push the card back to the Hub so it's better to fail early and not provide the option to fail later.
• push_to_hub do not accept **kwargs anymore. All accepted parameters are explicitly documented.
• arguments of push_to_hub are now keyword-only to avoid confusion. Only repo_id can be positional since it's the main arg.
• removed use_temp_dir argument from push_to_hub. We now use a tmp dir in all cases.

Linked PR: https://github.com/huggingface/transformers/pull/42391.

CLI

The deprecated transformers-cli ... command was deprecated, transformers ... is now the only CLI entry point.

transformers CLI has been migrated to Typer, making it easier to maintain + adding some nice features out of the box (improved --help section, autocompletion).

Biggest breaking change is in transformers chat. This command starts a terminal UI to interact with a chat model. It used to also be able to start a Chat Completion server powered by transformers and chat with it. In this revamped version, this feature has been removed in favor of transformers serve. The goal of splitting transformers chat and transformers serve is to define clear boundaries between client and server code. It helps with maintenance but also makes the commands less bloated. The new signature of transformers chat is:

Usage: transformers chat [OPTIONS] BASE_URL MODEL_ID [GENERATE_FLAGS]...

Chat with a model from the command line.

It works hand in hand with transformers serve, which means that if transformers serve is running on its default endpoint, transformers chat can be launched as follows:

transformers chat HuggingFaceTB/SmolLM3-3B

It can however use any OpenAI API compatible HTTP endpoint:

transformers chat HuggingFaceTB/SmolLM3-3B https://router.huggingface.co/v1

Linked PRs:

• https://github.com/huggingface/transformers/pull/40997
• https://github.com/huggingface/transformers/pull/41487

Removal of the run method

The transformers run (previously transformers-cli run) is an artefact of the past, was not documented nor tested, and isn't part of any public documentation. We're removing it for now and ask you to please let us know in case this is a method you are using; in which case we should bring it back with better support.

Linked PR: https://github.com/huggingface/transformers/pull/42447

Environment variables

• Legacy environment variables like TRANSFORMERS_CACHE, PYTORCH_TRANSFORMERS_CACHE, and PYTORCH_PRETRAINED_BERT_CACHE have been removed. Please use HF_HOME instead.
• Constants HUGGINGFACE_CO_EXAMPLES_TELEMETRY, HUGGINGFACE_CO_EXAMPLES_TELEMETRY, HUGGINGFACE_CO_PREFIX, and HUGGINGFACE_CO_RESOLVE_ENDPOINT have been removed. Please use huggingface_hub.constants.ENDPOINT instead.

Linked PR: https://github.com/huggingface/transformers/pull/42391.

Requirements update

transformers v5 pins the huggingface_hub version to >=1.0.0. See this migration guide to learn more about this major release. Here are to main aspects to know about:

• switched the HTTP backend from requests to httpx. This change was made to improve performance and to support both synchronous and asynchronous requests the same way. If you are currently catching requests.HTTPError errors in your codebase, you'll need to switch to httpx.HTTPError.
• related to 1., it is not possible to set proxies from your script. To handle proxies, you must set the HTTP_PROXY / HTTPS_PROXY environment variables
• hf_transfer and therefore HF_HUB_ENABLE_HF_TRANSFER have been completed dropped in favor of hf_xet. This should be transparent for most users. Please let us know if you notice any downside!

typer-slim has been added as required dependency, used to implement both hf and transformers CLIs.

New model additions in v5

CWM

The Code World Model (CWM) model was proposed in CWM: An Open-Weights LLM for Research on Code Generation with World Models by Meta FAIR CodeGen Team. CWM is an LLM for code generation and reasoning about code that has, in particular, been trained to better represent and reason about how code and commands affect the state of a program or system. Specifically, we mid-trained CWM on a large number of observation-action trajectories from Python execution traces and agentic interactions in containerized environments. We post-trained with extensive multi-task RL in verifiable coding, math, and multi-turn software engineering environments.

• Add Code World Model (CWM) by @jacobkahn in #41199

SAM3

SAM3 (Segment Anything Model 3) was introduced in SAM 3: Segment Anything with Concepts.

The SAM3 addition adds four new architectures:

• Sam3
• Sam3Tracker
• Sam3TrackerVideo
• Sam3Video

SAM3 performs Promptable Concept Segmentation (PCS) on images. PCS takes text and/or image exemplars as input (e.g., "yellow school bus"), and predicts instance and semantic masks for every single object matching the concept.

Sam3Tracker and Sam3TrackerVideo perform Promptable Visual Segmentation (PVS) on images. PVS takes interactive visual prompts (points, boxes, masks) or text inputs to segment a specific object instance per prompt. This is the task that SAM 1 and SAM 2 focused on, and SAM 3 improves upon it. Sam3Tracker and Sam3TrackerVideo are updated versions of SAM2 Video that maintain the same API while providing improved performance and capabilities.

SAM3 Video performs Promptable Concept Segmentation (PCS) on videos. PCS takes text as input (e.g., "yellow school bus"), and predicts instance and semantic masks for every single object matching the concept, while preserving object identities across video frames. The model combines a detection module (SAM3) with a tracking module (SAM2-style tracker) to enable robust object tracking across video frames using text prompts.

• Add SAM3 to 🤗 Transformers by @yonigozlan in #42285

LFM2 MoE

LFM2-MoE is a Mixture-of-Experts (MoE) variant of LFM2. The LFM2 family is optimized for on-device inference by combining short‑range, input‑aware gated convolutions with grouped‑query attention (GQA) in a layout tuned to maximize quality under strict speed and memory constraints.

LFM2‑MoE keeps this fast backbone and introduces sparse MoE feed‑forward networks to add representational capacity without significantly increasing the active compute path. The first LFM2-MoE release is LFM2-8B-A1B, with 8.3B total parameters and 1.5B active parameters. The model excels in quality (comparable to 3-4B dense models) and speed (faster than other 1.5B class models).

• [Model] Lfm2Moe by @paulpak58 in #41401

VideoLlama 3

The VideoLLaMA3 model is a major update to VideoLLaMA2 from Alibaba DAMO Academy.

• [model] Add VideoLLaMA3 implementation by @lkhl in #40499

AudioFlamingo 3

Audio Flamingo 3 (AF3) is a fully open large audio–language model designed for robust understanding and reasoning over speech, environmental sounds, and music. AF3 pairs a Whisper-style audio encoder with a causal language model and performs replace-in-place audio–text fusion: the processor aligns post-pool audio frames to a dedicated placeholder token and the model replaces those token slots with projected audio embeddings during the forward pass.

The model checkpoint is available at: nvidia/audio-flamingo-3-hf

Highlights:

• Unified audio encoder across speech, sound, and music.
• Long-audio support via windowing and post-pool alignment (up to 10 minutes maximum). The model processes audio in 30-second windows with a hard limit of 20 windows (10 minutes total). Audio longer than 10 minutes will be truncated.
• Deterministic fusion that preserves sequence length by replacing audio placeholder tokens with audio embeddings.

• [models] Add AudioFlamingo3 integration by @lashahub in #40290

Nanochat

NanoChat is a compact decoder-only transformer model designed for educational purposes and efficient training. The model features several fundamental architectural innovations which are common in modern transformer models. Therefore, it is a good model to use as a starting point to understand the principles of modern transformer models. NanoChat is a variant of the Llama architecture, with simplified attention mechanism and normalization layers.

• [MODEL] Nanochat implementation by @burtenshaw in #41634

FastVLM

FastVLM is an open-source vision-language model featuring a novel hybrid vision encoder, FastViTHD. Leveraging reparameterizable convolutional layers, scaled input resolution, and a reduced number of visual tokens, FastVLM delivers high accuracy with exceptional efficiency. Its optimized architecture enables deployment even on edge devices, achieving ultra-low TTFT (time to first token) without sacrificing performance.

• Add FastVLM by @camilla-deckard in https://github.com/huggingface/transformers/pull/41112

PaddleOCR-VL

PaddleOCR-VL is a SOTA and resource-efficient model tailored for document parsing. Its core component is PaddleOCR-VL-0.9B, a compact yet powerful vision-language model (VLM) that integrates a NaViT-style dynamic resolution visual encoder with the ERNIE-4.5-0.3B language model to enable accurate element recognition. This innovative model efficiently supports 109 languages and excels in recognizing complex elements (e.g., text, tables, formulas, and charts), while maintaining minimal resource consumption. Through comprehensive evaluations on widely used public benchmarks and in-house benchmarks, PaddleOCR-VL achieves SOTA performance in both page-level document parsing and element-level recognition. It significantly outperforms existing solutions, exhibits strong competitiveness against top-tier VLMs, and delivers fast inference speeds. These strengths make it highly suitable for practical deployment in real-world scenarios.

• [Model] Add PaddleOCR-VL Model Support by @zhang-prog in https://github.com/huggingface/transformers/pull/42178

SAM: Perception Encoder Audiovisual

PE Audio (Perception Encoder Audio) is a state-of-the-art multimodal model that embeds audio and text into a shared (joint) embedding space. The model enables cross-modal retrieval and understanding between audio and text.

Text input

• Produces a single embedding representing the full text.

Audio input

• PeAudioFrameLevelModel
  • Produces a sequence of embeddings, one every 40 ms of audio.
  • Suitable for audio event localization and fine-grained temporal analysis.
• PeAudioModel
  • Produces a single embedding for the entire audio clip.
  • Suitable for global audio-text retrieval tasks.

The resulting embeddings can be used for:

• Audio event localization
• Cross-modal (audio–text) retrieval and matching

• Sam: Perception Encoder Audiovisual by @eustlb in https://github.com/huggingface/transformers/pull/42905

Jais2

Jais2 a next-generation Arabic open-weight LLM trained on the richest Arabic-first dataset to date. Built from the ground up with 8B and 70B parameters, Jais 2 understands Arabic the way it's truly spoken across dialects, cuulutre, and modern expression. It is developed by MBZUAI, Inception and Cerebras Systems and based on the transformer architecture with modifications including:

• LayerNorm instead of RMSNorm
• ReLU² activation function
• Rotary Position Embeddings (RoPE)

• adds jais2 model support by @sarathc-cerebras in https://github.com/huggingface/transformers/pull/42684

Pixio

Pixio is a vision foundation model that uses ViT as a feature extractor for multiple downstream tasks like depth estimation, semantic segmentation, feed-forward 3D reconstruction, robotics, and image classification. It is built on the Masked Autoencoder (MAE) pre-training framework, with four minimal yet critical updates: 1) deeper decoder, 2) larger masking granularity, 3) more class tokens, and 4) web-scale curated training data.

• Add Pixio pre-trained models by @LiheYoung in https://github.com/huggingface/transformers/pull/42795

Ernie 4.5 VL MoE

The Ernie 4.5 VL MoE model was released in the Ernie 4.5 Model Family release by baidu. This family of models contains multiple different architectures and model sizes. The Vision-Language series in specific is composed of a novel multimodal heterogeneous structure, sharing paremeters across modalities and dedicating parameters to specific modalities. This becomes especially apparent in the Mixture of Expert (MoE) which is composed of

• Dedicated Text Experts
• Dedicated Vision Experts
• Shared Experts

This architecture has the advantage to enhance multimodal understanding without compromising, and even improving, performance on text-related tasks. An more detailed breakdown is given in the Technical Report.

• [Ernie 4.5] Ernie VL models by @vasqu in https://github.com/huggingface/transformers/pull/39585

GLM-ASR

GLM-ASR-Nano-2512 is a robust, open-source speech recognition model with 1.5B parameters. Designed for real-world complexity, it outperforms OpenAI Whisper V3 on multiple benchmarks while maintaining a compact size.

Key capabilities include:

• Exceptional Dialect Support Beyond standard Mandarin and English, the model is highly optimized for Cantonese (粤语) and other dialects, effectively bridging the gap in dialectal speech recognition.

• Low-Volume Speech Robustness Specifically trained for "Whisper/Quiet Speech" scenarios. It captures and accurately transcribes extremely low-volume audio that traditional models often miss.

• SOTA Performance Achieves the lowest average error rate (4.10) among comparable open-source models, showing significant advantages in Chinese benchmarks (Wenet Meeting, Aishell-1, etc..).

This model was contributed by Eustache Le Bihan and Yuxuan Zhang. you can check the model card for more details and our github repo.

• GLM-ASR Support by @zRzRzRzRzRzRzR in https://github.com/huggingface/transformers/pull/42875

GLM 4.7 Flash

GLM-4.7-Flash offers a new option for lightweight deployment that balances performance and efficiency.

• [GLM-4.7] GLM-Lite Support by @zRzRzRzRzRzRzR in https://github.com/huggingface/transformers/pull/43031

GLM Image

We present GLM-4.1V-Thinking, GLM-4.5V, and GLM-4.6V, a family of vision-language models (VLMs) designed to advance general-purpose multimodal understanding and reasoning. In this report, we share our key findings in the development of the reasoning-centric training framework. We first develop a capable vision foundation model with significant potential through large-scale pre-training, which arguably sets the upper bound for the final performance. We then propose Reinforcement Learning with Curriculum Sampling (RLCS) to unlock the full potential of the model, leading to comprehensive capability enhancement across a diverse range of tasks, including STEM problem solving, video understanding, content recognition, coding, grounding, GUI-based agents, and long document interpretation. In a comprehensive evaluation across 42 public benchmarks, GLM-4.5V achieves state-of-the-art performance on nearly all tasks among open-source models of similar size, and demonstrates competitive or even superior results compared to closed-source models such as Gemini-2.5-Flash on challenging tasks including Coding and GUI Agents. Meanwhile, the smaller GLM-4.1V-9B-Thinking remains highly competitive-achieving superior results to the much larger Qwen2.5-VL-72B on 29 benchmarks. We open-source both GLM-4.1V-9B-Thinking and GLM-4.5V. We further introduce the GLM-4.6V series, open-source multimodal models with native tool use and a 128K context window. A brief overview is available at this https URL. Code, models and more information are released at https://github.com/zai-org/GLM-V

• [GLM-Image] AR Model Support for GLM-Image by @zRzRzRzRzRzRzR in https://github.com/huggingface/transformers/pull/43100

LWDetr

LW-DETR proposes a light-weight Detection Transformer (DETR) architecture designed to compete with and surpass the dominant YOLO series for real-time object detection. It achieves a new state-of-the-art balance between speed (latency) and accuracy (mAP) by combining recent transformer advances with efficient design choices.

The LW-DETR architecture is characterized by its simple and efficient structure: a plain ViT Encoder, a Projector, and a shallow DETR Decoder. It enhances the DETR architecture for efficiency and speed using the following core modifications:

Efficient ViT Encoder: Uses a plain ViT with interleaved window/global attention and a window-major organization to drastically reduce attention complexity and latency.
Richer Input: Aggregates multi-level features from the encoder and uses a C2f Projector (YOLOv8) to pass two-scale features ( 1 / 8 and 1 / 32 ).
Faster Decoder: Employs a shallow 3-layer DETR decoder with deformable cross-attention for lower latency and faster convergence.
Optimized Queries: Uses a mixed-query scheme combining learnable content queries and generated spatial queries.

• Add LWDetr model by @sbucaille in https://github.com/huggingface/transformers/pull/40991

LightOnOCR

LightOnOcr combines a Vision Transformer encoder (Pixtral-based) with a lightweight text decoder (Qwen3-based) distilled from high-quality open VLMs. It is optimized for document parsing tasks, producing accurate, layout-aware text extraction from high-resolution pages.

• Add LightOnOCR model implementation by @baptiste-aubertin in https://github.com/huggingface/transformers/pull/41621

Bugfixes and improvements

• JetMoe Fix jetmoe after #40132 by @ArthurZucker in #41324
• Fixed tiny incorrect import in gemma3 by @Sai-Suraj-27 in #41354
• Rope for Qwen2--5-vl by @zucchini-nlp in #41173
• 🚨 Bump to Python 3.10 and rework how we check 3rd-party libraries existence by @Cyrilvallez in #41268
• Standardize PretrainedConfig to PreTrainedConfig by @Cyrilvallez in #41300
• Fix trainer for py3.9 by @SunMarc in #41359
• Check model inputs - hidden states by @zucchini-nlp in #40994
• [ModularChecker] QOL for the modular checker by @ArthurZucker in #41361
• Fixing a typo for BLT model by @Narsil in #41325
• :rotating_light: [v5] Remove relative position embeddings (for bert like models) by @vasqu in #41170
• Fix typo in model proposal template by @Ombucha in #41352
• Better typehints for apply_chat_template by @Samoed in #41355
• 🚨 Remove BetterTransformer by @Cyrilvallez in #41367
• [testing] update test_longcat_generation_cpu by @ydshieh in #41368
• Fix flash_attention.py: wrong argument passing for attn_implementation by @TKONIY in #41347
• Use canonical get_size_with_aspect_ratio (with max_size) from transformers.image_transforms to fix #37939 by @sonianuj287 in #41284
• Fixes in check_model_inputs, GPTBigCodeModel and ImageGPTModel by @IlyasMoutawwakil in #40811
• Remove unnecessary list comprehension by @cyyever in #41305
• make some ut cases pass on xpu w/ latest torch by @yao-matrix in #41337
• Remove unused function patameters by @cyyever in #41358
• [CB] Refactors the way we access paged by @ArthurZucker in #41370
• serve: add non-streaming mode to /v1/responses; stream event parity; remove placeholder logprobs by @antznette1 in #41353
• Update from pretrained error when loading by @ArthurZucker in #33380
• [v5] Sync Bert and Bart eager attention by @vasqu in #41248
• fix asr ut failures by @yao-matrix in #41332
• fix resample in asr pipeline by @yhzx233 in #41298
• Correct numerical regression in vision embeddings by @i3hz in #41374
• [kernels] Kernel Config by @MekkCyber in #41232
• [Cache] lfm2 cache: allocate empty kv layers during init by @paulpak58 in #41396
• Fix test for model with dotted name and relative imports by @st81 in #41343
• Prefer raising TypeError exception for invalid type by @Sai-Suraj-27 in #41346

Significant community contributions

The following contributors have made significant changes to the library over the last release:

• @ArthurZucker
  • JetMoe Fix jetmoe after #40132 (#41324)
  • [ModularChecker] QOL for the modular checker (#41361)
  • [CB] Refactors the way we access paged (#41370)
  • Update from pretrained error when loading (#33380)
  • :facepalm: CB nit! (#41413)
  • [from_pretrained] Small refactor from_pretrained: move around unrelated stuff (#41445)
  • update deps table (#42120)
  • Refactor weight loading (#41580)
  • Update conversion mapping to separate renaming from converting (#42254)
  • Auto convert tekken.json (#42299)
  • fix tekken pattern matching (#42363)
  • Small tp fix (#42366)
  • Fix tp (#42368)
  • misc don't recreate it (#42394)
• @vasqu
  • :rotating_light: [v5] Remove relative position embeddings (for bert like models) (#41170)
  • [v5] Sync Bert and Bart eager attention (#41248)
  • [JetMoe] Fix KV head repetition and padding free (#41423)
  • :rotating_light: [Attention Masks] Bidirectional masks for encoder and encoder-decoder models (#41265)
  • [CI] Fix copies on main (#41486)
  • [Docs] Fix changed references (#41614)
  • [Executorch] Simplify for encoder models (#41627)
  • [Ernie 4.5 Moe] Fix Moe and offloading (#41385)
  • [Masks] Fix mask handling in eager for vision models (#41625)
  • [Attn] Allow dynamic causality in SDPA via Kwargs (#41692)
  • [Onnx docs] Remove some traces (#41791)
  • :rotating_light: [Clip] Fix masking and enable flash attention on all model types (#41750)
  • [Attn Masks] Non-vmap default for attention masks (#41852)
  • [T5Gemma] Fix cross attention cache (#41890)
  • [Pop2Piano] Fix cache usage (#42170)
  • [PEFT] Fix prefix tuning (#41696)
  • [PEFT] Fix the general test for prefix tuning (#42185)
  • [Pop2Piano] Fix tied weights (#42193)
  • [BLT] Fix cache usage (#42188)
  • [CI] Skip EfficientLoFTR test (#42327)
  • [Attn Masks] Lift bidirectional mask restriction on eager (#42325)