[Add Mixtral] Adds support for the Mixtral MoE (#27942)
* up * up * test * logits ok * up * up * few fixes * conversion script * up * nits * nits * update * nuke * more updates * nites * fix many issues * nit * scatter * nit * nuke megablocks * nits * fix conversion script * nit * remove * nits * nit * update * oupsssss * change * nits device * nits * fixup * update * merge * add copied from * fix the copy mentions * update tests * more fixes * nits * conversion script * add parts of the readme * Update tests/models/mixtral/test_modeling_mixtral.py Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com> * new test + conversion script * Apply suggestions from code review Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com> * Apply suggestions from code review * fix * fix copies * fix copies * ooops * fix config * Apply suggestions from code review * fix nits * nit * add copies * add batched tests * docs * fix flash attention * let's add more verbose * add correct outputs * support router ouptus * ignore copies where needed * fix * cat list if list is given for now * nits * Update docs/source/en/model_doc/mixtral.md * finish router refactoring * fix forward * fix expected values * nits * fixup * fix * fix bug * fix * fix dtype mismatch * fix * grrr grrr I support item assignment * fix CI * docs * fixup * remove some copied form * fix weird diff * skip doctest fast on the config and modeling * mark that is supports flash attention in the doc * update * Update src/transformers/models/mixtral/modeling_mixtral.py Co-authored-by: Lysandre Debut <hi@lysand.re> * Update docs/source/en/model_doc/mixtral.md Co-authored-by: Lysandre Debut <hi@lysand.re> * revert router logits config issue * update doc accordingly * Update src/transformers/models/mixtral/convert_mixtral_weights_to_hf.py * nits * use torch testing asssert close * fixup * doc nits --------- Co-authored-by: younesbelkada <younesbelkada@gmail.com> Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com> Co-authored-by: Lysandre Debut <hi@lysand.re>
This commit is contained in:
Arthur
@@ -396,6 +396,8 @@
|
||||
title: MegatronGPT2
|
||||
- local: model_doc/mistral
|
||||
title: Mistral
|
||||
- local: model_doc/mixtral
|
||||
title: Mixtral
|
||||
- local: model_doc/mluke
|
||||
title: mLUKE
|
||||
- local: model_doc/mobilebert
|
||||
|
||||
@@ -189,6 +189,7 @@ Flax), PyTorch, and/or TensorFlow.
|
||||
| [Megatron-GPT2](model_doc/megatron_gpt2) | ✅ | ✅ | ✅ |
|
||||
| [MGP-STR](model_doc/mgp-str) | ✅ | ❌ | ❌ |
|
||||
| [Mistral](model_doc/mistral) | ✅ | ❌ | ❌ |
|
||||
| [Mixtral](model_doc/mixtral) | ✅ | ❌ | ❌ |
|
||||
| [mLUKE](model_doc/mluke) | ✅ | ❌ | ❌ |
|
||||
| [MMS](model_doc/mms) | ✅ | ✅ | ✅ |
|
||||
| [MobileBERT](model_doc/mobilebert) | ✅ | ✅ | ❌ |
|
||||
|
||||
163
docs/source/en/model_doc/mixtral.md
Normal file
163
docs/source/en/model_doc/mixtral.md
Normal file
@@ -0,0 +1,163 @@
|
||||
<!--Copyright 2023 Mistral AI and The HuggingFace Team. All rights reserved.
|
||||
|
||||
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
|
||||
the License. You may obtain a copy of the License at
|
||||
|
||||
http://www.apache.org/licenses/LICENSE-2.0
|
||||
|
||||
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
|
||||
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
|
||||
specific language governing permissions and limitations under the License.
|
||||
|
||||
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
|
||||
rendered properly in your Markdown viewer.
|
||||
|
||||
-->
|
||||
|
||||
# Mixtral
|
||||
|
||||
## Overview
|
||||
|
||||
Mixtral-8x7B is Mistral AI's second Large Language Model (LLM).
|
||||
|
||||
The Mixtral model was proposed in the by the [Mistral AI](https://mistral.ai/) team.
|
||||
|
||||
It was introduced in the [Mixtral of Experts blogpost](https://mistral.ai/news/mixtral-of-experts/) with the following introduction:
|
||||
|
||||
*Today, the team is proud to release Mixtral 8x7B, a high-quality sparse mixture of experts models (SMoE) with open weights. Licensed under Apache 2.0. Mixtral outperforms Llama 2 70B on most benchmarks with 6x faster inference. It is the strongest open-weight model with a permissive license and the best model overall regarding cost/performance trade-offs. In particular, it matches or outperforms GPT3.5 on most standard benchmarks.*
|
||||
|
||||
Tips:
|
||||
|
||||
|
||||
- The model needs to be converted using the [conversion script](https://github.com/huggingface/transformers/blob/main/src/transformers/models/mixtral/convert_mixtral_weights_to_hf.py).
|
||||
- If the model is quantized to 4bits, a single A100 is enough to fit the entire 84B model.
|
||||
|
||||
This model was contributed by [Younes Belkada](https://huggingface.co/ybelkada) and [Arthur Zucker](https://huggingface.co/ArthurZ) .
|
||||
The original code can be found [here](https://github.com/mistralai/mistral-src).
|
||||
|
||||
|
||||
### Model Details
|
||||
|
||||
Mixtral-84B is a decoder-based LM with the following architectural choices:
|
||||
|
||||
* Mixtral is a Mixture of Expert (MOE) model with 8 experts per MLP, with a total of 85B paramateres but the compute required is the same as a 14B model. This is because even though each experts have to be loaded in RAM (70B like ram requirement) each token from the hidden states are dipatched twice (top 2 routing) and thus the compute (the operation required at each foward computation) is just 2 X sequence_length.
|
||||
|
||||
The following implementation details are shared with Mistral AI's first model [mistral](~models/doc/mistral):
|
||||
* Sliding Window Attention - Trained with 8k context length and fixed cache size, with a theoretical attention span of 128K tokens
|
||||
* GQA (Grouped Query Attention) - allowing faster inference and lower cache size.
|
||||
* Byte-fallback BPE tokenizer - ensures that characters are never mapped to out of vocabulary tokens.
|
||||
|
||||
They also provide an instruction fine-tuned model: `mistralai/Mixtral-8x7B-v0.1` which can be used for chat-based inference.
|
||||
|
||||
For more details please read our [release blog post](https://mistral.ai/news/mixtral-of-experts/)
|
||||
|
||||
### License
|
||||
|
||||
`Mixtral-8x7B` is released under the Apache 2.0 license.
|
||||
|
||||
## Usage tips
|
||||
|
||||
`Mixtral-8x7B` can be found on the [Huggingface Hub](https://huggingface.co/mistralai)
|
||||
|
||||
These ready-to-use checkpoints can be downloaded and used via the HuggingFace Hub:
|
||||
|
||||
```python
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
>>> device = "cuda" # the device to load the model onto
|
||||
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("mistralai/Mixtral-8x7B-v0.1")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-8x7B")
|
||||
|
||||
>>> prompt = "My favourite condiment is"
|
||||
|
||||
>>> model_inputs = tokenizer([prompt], return_tensors="pt").to(device)
|
||||
>>> model.to(device)
|
||||
|
||||
>>> generated_ids = model.generate(**model_inputs, max_new_tokens=100, do_sample=True)
|
||||
>>> tokenizer.batch_decode(generated_ids)[0]
|
||||
"The expected output"
|
||||
```
|
||||
|
||||
To use the raw checkpoints with HuggingFace you can use the `convert_mixtral_weights_to_hf.py` script to convert them to the HuggingFace format:
|
||||
|
||||
```bash
|
||||
python src/transformers/models/mixtral/convert_mixtral_weights_to_hf.py \
|
||||
--input_dir /path/to/downloaded/mistral/weights --output_dir /output/path
|
||||
```
|
||||
|
||||
You can then load the converted model from the `output/path`:
|
||||
|
||||
```python
|
||||
from transformers import MixtralForCausalLM, LlamaTokenizer
|
||||
|
||||
tokenizer = LlamaTokenizer.from_pretrained("/output/path")
|
||||
model = MixtralForCausalLM.from_pretrained("/output/path")
|
||||
```
|
||||
|
||||
## Combining Mixtral and Flash Attention 2
|
||||
|
||||
First, make sure to install the latest version of Flash Attention 2 to include the sliding window attention feature.
|
||||
|
||||
```bash
|
||||
pip install -U flash-attn --no-build-isolation
|
||||
```
|
||||
|
||||
Make also sure that you have a hardware that is compatible with Flash-Attention 2. Read more about it in the official documentation of [`flash-attn`](https://github.com/Dao-AILab/flash-attention) repository. Make also sure to load your model in half-precision (e.g. `torch.float16`)
|
||||
|
||||
To load and run a model using Flash Attention 2, refer to the snippet below:
|
||||
|
||||
```python
|
||||
>>> import torch
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
>>> device = "cuda" # the device to load the model onto
|
||||
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("mistralai/Mixtral-8x7B-v0.1", torch_dtype=torch.float16, attn_implementation="flash_attention_2")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("mistralai/Mixtral-8x7B-v0.1")
|
||||
|
||||
>>> prompt = "My favourite condiment is"
|
||||
|
||||
>>> model_inputs = tokenizer([prompt], return_tensors="pt").to(device)
|
||||
>>> model.to(device)
|
||||
|
||||
>>> generated_ids = model.generate(**model_inputs, max_new_tokens=100, do_sample=True)
|
||||
>>> tokenizer.batch_decode(generated_ids)[0]
|
||||
"The expected output"
|
||||
```
|
||||
|
||||
### Expected speedups
|
||||
|
||||
Below is a expected speedup diagram that compares pure inference time between the native implementation in transformers using `mistralai/Mixtral-8x7B-v0.1` checkpoint and the Flash Attention 2 version of the model.
|
||||
|
||||
<div style="text-align: center">
|
||||
<img src="https://huggingface.co/datasets/ybelkada/documentation-images/resolve/main/mixtral-7b-inference-large-seqlen.png">
|
||||
</div>
|
||||
|
||||
### Sliding window Attention
|
||||
|
||||
The current implementation supports the sliding window attention mechanism and memory efficient cache management.
|
||||
To enable sliding window attention, just make sure to have a `flash-attn` version that is compatible with sliding window attention (`>=2.3.0`).
|
||||
|
||||
The Flash Attention-2 model uses also a more memory efficient cache slicing mechanism - as recommended per the official implementation of Mistral model that use rolling cache mechanism we keep the cache size fixed (`self.config.sliding_window`), support batched generation only for `padding_side="left"` and use the absolute position of the current token to compute the positional embedding.
|
||||
|
||||
## The Mistral Team
|
||||
|
||||
Albert Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lélio Renard Lavaud, Lucile Saulnier, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix, William El Sayed.
|
||||
|
||||
## MixtralConfig
|
||||
|
||||
[[autodoc]] MixtralConfig
|
||||
|
||||
## MixtralModel
|
||||
|
||||
[[autodoc]] MixtralModel
|
||||
- forward
|
||||
|
||||
## MixtralForCausalLM
|
||||
|
||||
[[autodoc]] MixtralForCausalLM
|
||||
- forward
|
||||
|
||||
## MixtralForSequenceClassification
|
||||
|
||||
[[autodoc]] MixtralForSequenceClassification
|
||||
- forward
|
||||
@@ -48,6 +48,7 @@ FlashAttention-2 is currently supported for the following architectures:
|
||||
* [Llava](https://huggingface.co/docs/transformers/model_doc/llava)
|
||||
* [MBart](https://huggingface.co/docs/transformers/model_doc/mbart#transformers.MBartModel)
|
||||
* [Mistral](https://huggingface.co/docs/transformers/model_doc/mistral#transformers.MistralModel)
|
||||
* [Mixtral](https://huggingface.co/docs/transformers/model_doc/mixtral#transformers.MixtralModel)
|
||||
* [OPT](https://huggingface.co/docs/transformers/model_doc/opt#transformers.OPTModel)
|
||||
* [Phi](https://huggingface.co/docs/transformers/model_doc/phi#transformers.PhiModel)
|
||||
* [Whisper](https://huggingface.co/docs/transformers/model_doc/whisper#transformers.WhisperModel)
|
||||
|
||||
@@ -37,7 +37,7 @@ You can finetune other architectures for causal language modeling following the
|
||||
Choose one of the following architectures:
|
||||
|
||||
<!--This tip is automatically generated by `make fix-copies`, do not fill manually!-->
|
||||
[BART](../model_doc/bart), [BERT](../model_doc/bert), [Bert Generation](../model_doc/bert-generation), [BigBird](../model_doc/big_bird), [BigBird-Pegasus](../model_doc/bigbird_pegasus), [BioGpt](../model_doc/biogpt), [Blenderbot](../model_doc/blenderbot), [BlenderbotSmall](../model_doc/blenderbot-small), [BLOOM](../model_doc/bloom), [CamemBERT](../model_doc/camembert), [CodeLlama](../model_doc/code_llama), [CodeGen](../model_doc/codegen), [CPM-Ant](../model_doc/cpmant), [CTRL](../model_doc/ctrl), [Data2VecText](../model_doc/data2vec-text), [ELECTRA](../model_doc/electra), [ERNIE](../model_doc/ernie), [Falcon](../model_doc/falcon), [Fuyu](../model_doc/fuyu), [GIT](../model_doc/git), [GPT-Sw3](../model_doc/gpt-sw3), [OpenAI GPT-2](../model_doc/gpt2), [GPTBigCode](../model_doc/gpt_bigcode), [GPT Neo](../model_doc/gpt_neo), [GPT NeoX](../model_doc/gpt_neox), [GPT NeoX Japanese](../model_doc/gpt_neox_japanese), [GPT-J](../model_doc/gptj), [LLaMA](../model_doc/llama), [Marian](../model_doc/marian), [mBART](../model_doc/mbart), [MEGA](../model_doc/mega), [Megatron-BERT](../model_doc/megatron-bert), [Mistral](../model_doc/mistral), [MPT](../model_doc/mpt), [MusicGen](../model_doc/musicgen), [MVP](../model_doc/mvp), [OpenLlama](../model_doc/open-llama), [OpenAI GPT](../model_doc/openai-gpt), [OPT](../model_doc/opt), [Pegasus](../model_doc/pegasus), [Persimmon](../model_doc/persimmon), [Phi](../model_doc/phi), [PLBart](../model_doc/plbart), [ProphetNet](../model_doc/prophetnet), [QDQBert](../model_doc/qdqbert), [Reformer](../model_doc/reformer), [RemBERT](../model_doc/rembert), [RoBERTa](../model_doc/roberta), [RoBERTa-PreLayerNorm](../model_doc/roberta-prelayernorm), [RoCBert](../model_doc/roc_bert), [RoFormer](../model_doc/roformer), [RWKV](../model_doc/rwkv), [Speech2Text2](../model_doc/speech_to_text_2), [Transformer-XL](../model_doc/transfo-xl), [TrOCR](../model_doc/trocr), [Whisper](../model_doc/whisper), [XGLM](../model_doc/xglm), [XLM](../model_doc/xlm), [XLM-ProphetNet](../model_doc/xlm-prophetnet), [XLM-RoBERTa](../model_doc/xlm-roberta), [XLM-RoBERTa-XL](../model_doc/xlm-roberta-xl), [XLNet](../model_doc/xlnet), [X-MOD](../model_doc/xmod)
|
||||
[BART](../model_doc/bart), [BERT](../model_doc/bert), [Bert Generation](../model_doc/bert-generation), [BigBird](../model_doc/big_bird), [BigBird-Pegasus](../model_doc/bigbird_pegasus), [BioGpt](../model_doc/biogpt), [Blenderbot](../model_doc/blenderbot), [BlenderbotSmall](../model_doc/blenderbot-small), [BLOOM](../model_doc/bloom), [CamemBERT](../model_doc/camembert), [CodeLlama](../model_doc/code_llama), [CodeGen](../model_doc/codegen), [CPM-Ant](../model_doc/cpmant), [CTRL](../model_doc/ctrl), [Data2VecText](../model_doc/data2vec-text), [ELECTRA](../model_doc/electra), [ERNIE](../model_doc/ernie), [Falcon](../model_doc/falcon), [Fuyu](../model_doc/fuyu), [GIT](../model_doc/git), [GPT-Sw3](../model_doc/gpt-sw3), [OpenAI GPT-2](../model_doc/gpt2), [GPTBigCode](../model_doc/gpt_bigcode), [GPT Neo](../model_doc/gpt_neo), [GPT NeoX](../model_doc/gpt_neox), [GPT NeoX Japanese](../model_doc/gpt_neox_japanese), [GPT-J](../model_doc/gptj), [LLaMA](../model_doc/llama), [Marian](../model_doc/marian), [mBART](../model_doc/mbart), [MEGA](../model_doc/mega), [Megatron-BERT](../model_doc/megatron-bert), [Mistral](../model_doc/mistral), [Mixtral](../model_doc/mixtral), [MPT](../model_doc/mpt), [MusicGen](../model_doc/musicgen), [MVP](../model_doc/mvp), [OpenLlama](../model_doc/open-llama), [OpenAI GPT](../model_doc/openai-gpt), [OPT](../model_doc/opt), [Pegasus](../model_doc/pegasus), [Persimmon](../model_doc/persimmon), [Phi](../model_doc/phi), [PLBart](../model_doc/plbart), [ProphetNet](../model_doc/prophetnet), [QDQBert](../model_doc/qdqbert), [Reformer](../model_doc/reformer), [RemBERT](../model_doc/rembert), [RoBERTa](../model_doc/roberta), [RoBERTa-PreLayerNorm](../model_doc/roberta-prelayernorm), [RoCBert](../model_doc/roc_bert), [RoFormer](../model_doc/roformer), [RWKV](../model_doc/rwkv), [Speech2Text2](../model_doc/speech_to_text_2), [Transformer-XL](../model_doc/transfo-xl), [TrOCR](../model_doc/trocr), [Whisper](../model_doc/whisper), [XGLM](../model_doc/xglm), [XLM](../model_doc/xlm), [XLM-ProphetNet](../model_doc/xlm-prophetnet), [XLM-RoBERTa](../model_doc/xlm-roberta), [XLM-RoBERTa-XL](../model_doc/xlm-roberta-xl), [XLNet](../model_doc/xlnet), [X-MOD](../model_doc/xmod)
|
||||
|
||||
|
||||
|
||||
|
||||
@@ -33,7 +33,7 @@ The task illustrated in this tutorial is supported by the following model archit
|
||||
<!--This tip is automatically generated by `make fix-copies`, do not fill manually!-->
|
||||
|
||||
|
||||
[ALBERT](../model_doc/albert), [BART](../model_doc/bart), [BERT](../model_doc/bert), [BigBird](../model_doc/big_bird), [BigBird-Pegasus](../model_doc/bigbird_pegasus), [BioGpt](../model_doc/biogpt), [BLOOM](../model_doc/bloom), [CamemBERT](../model_doc/camembert), [CANINE](../model_doc/canine), [CodeLlama](../model_doc/code_llama), [ConvBERT](../model_doc/convbert), [CTRL](../model_doc/ctrl), [Data2VecText](../model_doc/data2vec-text), [DeBERTa](../model_doc/deberta), [DeBERTa-v2](../model_doc/deberta-v2), [DistilBERT](../model_doc/distilbert), [ELECTRA](../model_doc/electra), [ERNIE](../model_doc/ernie), [ErnieM](../model_doc/ernie_m), [ESM](../model_doc/esm), [Falcon](../model_doc/falcon), [FlauBERT](../model_doc/flaubert), [FNet](../model_doc/fnet), [Funnel Transformer](../model_doc/funnel), [GPT-Sw3](../model_doc/gpt-sw3), [OpenAI GPT-2](../model_doc/gpt2), [GPTBigCode](../model_doc/gpt_bigcode), [GPT Neo](../model_doc/gpt_neo), [GPT NeoX](../model_doc/gpt_neox), [GPT-J](../model_doc/gptj), [I-BERT](../model_doc/ibert), [LayoutLM](../model_doc/layoutlm), [LayoutLMv2](../model_doc/layoutlmv2), [LayoutLMv3](../model_doc/layoutlmv3), [LED](../model_doc/led), [LiLT](../model_doc/lilt), [LLaMA](../model_doc/llama), [Longformer](../model_doc/longformer), [LUKE](../model_doc/luke), [MarkupLM](../model_doc/markuplm), [mBART](../model_doc/mbart), [MEGA](../model_doc/mega), [Megatron-BERT](../model_doc/megatron-bert), [Mistral](../model_doc/mistral), [MobileBERT](../model_doc/mobilebert), [MPNet](../model_doc/mpnet), [MPT](../model_doc/mpt), [MRA](../model_doc/mra), [MT5](../model_doc/mt5), [MVP](../model_doc/mvp), [Nezha](../model_doc/nezha), [Nyströmformer](../model_doc/nystromformer), [OpenLlama](../model_doc/open-llama), [OpenAI GPT](../model_doc/openai-gpt), [OPT](../model_doc/opt), [Perceiver](../model_doc/perceiver), [Persimmon](../model_doc/persimmon), [Phi](../model_doc/phi), [PLBart](../model_doc/plbart), [QDQBert](../model_doc/qdqbert), [Reformer](../model_doc/reformer), [RemBERT](../model_doc/rembert), [RoBERTa](../model_doc/roberta), [RoBERTa-PreLayerNorm](../model_doc/roberta-prelayernorm), [RoCBert](../model_doc/roc_bert), [RoFormer](../model_doc/roformer), [SqueezeBERT](../model_doc/squeezebert), [T5](../model_doc/t5), [TAPAS](../model_doc/tapas), [Transformer-XL](../model_doc/transfo-xl), [UMT5](../model_doc/umt5), [XLM](../model_doc/xlm), [XLM-RoBERTa](../model_doc/xlm-roberta), [XLM-RoBERTa-XL](../model_doc/xlm-roberta-xl), [XLNet](../model_doc/xlnet), [X-MOD](../model_doc/xmod), [YOSO](../model_doc/yoso)
|
||||
[ALBERT](../model_doc/albert), [BART](../model_doc/bart), [BERT](../model_doc/bert), [BigBird](../model_doc/big_bird), [BigBird-Pegasus](../model_doc/bigbird_pegasus), [BioGpt](../model_doc/biogpt), [BLOOM](../model_doc/bloom), [CamemBERT](../model_doc/camembert), [CANINE](../model_doc/canine), [CodeLlama](../model_doc/code_llama), [ConvBERT](../model_doc/convbert), [CTRL](../model_doc/ctrl), [Data2VecText](../model_doc/data2vec-text), [DeBERTa](../model_doc/deberta), [DeBERTa-v2](../model_doc/deberta-v2), [DistilBERT](../model_doc/distilbert), [ELECTRA](../model_doc/electra), [ERNIE](../model_doc/ernie), [ErnieM](../model_doc/ernie_m), [ESM](../model_doc/esm), [Falcon](../model_doc/falcon), [FlauBERT](../model_doc/flaubert), [FNet](../model_doc/fnet), [Funnel Transformer](../model_doc/funnel), [GPT-Sw3](../model_doc/gpt-sw3), [OpenAI GPT-2](../model_doc/gpt2), [GPTBigCode](../model_doc/gpt_bigcode), [GPT Neo](../model_doc/gpt_neo), [GPT NeoX](../model_doc/gpt_neox), [GPT-J](../model_doc/gptj), [I-BERT](../model_doc/ibert), [LayoutLM](../model_doc/layoutlm), [LayoutLMv2](../model_doc/layoutlmv2), [LayoutLMv3](../model_doc/layoutlmv3), [LED](../model_doc/led), [LiLT](../model_doc/lilt), [LLaMA](../model_doc/llama), [Longformer](../model_doc/longformer), [LUKE](../model_doc/luke), [MarkupLM](../model_doc/markuplm), [mBART](../model_doc/mbart), [MEGA](../model_doc/mega), [Megatron-BERT](../model_doc/megatron-bert), [Mistral](../model_doc/mistral), [Mixtral](../model_doc/mixtral), [MobileBERT](../model_doc/mobilebert), [MPNet](../model_doc/mpnet), [MPT](../model_doc/mpt), [MRA](../model_doc/mra), [MT5](../model_doc/mt5), [MVP](../model_doc/mvp), [Nezha](../model_doc/nezha), [Nyströmformer](../model_doc/nystromformer), [OpenLlama](../model_doc/open-llama), [OpenAI GPT](../model_doc/openai-gpt), [OPT](../model_doc/opt), [Perceiver](../model_doc/perceiver), [Persimmon](../model_doc/persimmon), [Phi](../model_doc/phi), [PLBart](../model_doc/plbart), [QDQBert](../model_doc/qdqbert), [Reformer](../model_doc/reformer), [RemBERT](../model_doc/rembert), [RoBERTa](../model_doc/roberta), [RoBERTa-PreLayerNorm](../model_doc/roberta-prelayernorm), [RoCBert](../model_doc/roc_bert), [RoFormer](../model_doc/roformer), [SqueezeBERT](../model_doc/squeezebert), [T5](../model_doc/t5), [TAPAS](../model_doc/tapas), [Transformer-XL](../model_doc/transfo-xl), [UMT5](../model_doc/umt5), [XLM](../model_doc/xlm), [XLM-RoBERTa](../model_doc/xlm-roberta), [XLM-RoBERTa-XL](../model_doc/xlm-roberta-xl), [XLNet](../model_doc/xlnet), [X-MOD](../model_doc/xmod), [YOSO](../model_doc/yoso)
|
||||
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user