xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-07-07 13:27:23 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Doc] Update Qwen documentation (#9869)

Browse Source
This commit is contained in:
Jee Jee Li 2024-10-31 16:54:18 +08:00 committed by GitHub
parent 3ea2dc2ec4
commit 5608e611c2
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 6 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
docs/source/models/supported_models.rst
View File

@ -277,7 +277,7 @@ Text Generation
* - :code:`QWenLMHeadModel` * - :code:`QWenLMHeadModel`
- Qwen - Qwen
- :code:`Qwen/Qwen-7B`, :code:`Qwen/Qwen-7B-Chat`, etc. - :code:`Qwen/Qwen-7B`, :code:`Qwen/Qwen-7B-Chat`, etc.
- - ✅︎
- ✅︎ - ✅︎
* - :code:`Qwen2ForCausalLM` * - :code:`Qwen2ForCausalLM`
- Qwen2 - Qwen2
@ -516,7 +516,7 @@ Text Generation
- Qwen-VL - Qwen-VL
- T + I\ :sup:`E+` - T + I\ :sup:`E+`
- :code:`Qwen/Qwen-VL`, :code:`Qwen/Qwen-VL-Chat`, etc. - :code:`Qwen/Qwen-VL`, :code:`Qwen/Qwen-VL-Chat`, etc.
- - ✅︎
- ✅︎ - ✅︎
* - :code:`Qwen2AudioForConditionalGeneration` * - :code:`Qwen2AudioForConditionalGeneration`
- Qwen2-Audio - Qwen2-Audio
@ -540,6 +540,9 @@ Text Generation
| :sup:`E` Pre-computed embeddings can be inputted for this modality. | :sup:`E` Pre-computed embeddings can be inputted for this modality.
| :sup:`+` Multiple items can be inputted per text prompt for this modality. | :sup:`+` Multiple items can be inputted per text prompt for this modality.
.. note::
vLLM currently only supports adding LoRA to the language backbone of multimodal models.
.. note:: .. note::
For :code:`openbmb/MiniCPM-V-2`, the official repo doesn't work yet, so we need to use a fork (:code:`HwwwH/MiniCPM-V-2`) for now. For :code:`openbmb/MiniCPM-V-2`, the official repo doesn't work yet, so we need to use a fork (:code:`HwwwH/MiniCPM-V-2`) for now.
For more details, please see: https://github.com/vllm-project/vllm/pull/4087#issuecomment-2250397630 For more details, please see: https://github.com/vllm-project/vllm/pull/4087#issuecomment-2250397630

2
vllm/model_executor/models/qwen.py
View File

@ -1048,7 +1048,7 @@ class QWenVL(QWenBaseModel):
@MULTIMODAL_REGISTRY.register_max_image_tokens(MAX_QWEN_IMG_TOKENS) @MULTIMODAL_REGISTRY.register_max_image_tokens(MAX_QWEN_IMG_TOKENS)
@INPUT_REGISTRY.register_dummy_data(dummy_data_for_qwen) @INPUT_REGISTRY.register_dummy_data(dummy_data_for_qwen)
@INPUT_REGISTRY.register_input_processor(input_processor_for_qwen) @INPUT_REGISTRY.register_input_processor(input_processor_for_qwen)
class QWenLMHeadModel(QWenBaseModel): class QWenLMHeadModel(QWenBaseModel, SupportsLoRA):
""" """
QWenLMHeadModel is not only applicable to LLM but also to VL, which is not QWenLMHeadModel is not only applicable to LLM but also to VL, which is not
conducive to the current integration logic of LoRA in vLLM. Therefore, it conducive to the current integration logic of LoRA in vLLM. Therefore, it