biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

rename PromptInputs and inputs with backward compatibility (#8760)

Browse Source
This commit is contained in:
Cyrus Leung
2024-09-26 00:36:47 +08:00
committed by GitHub
parent 0c4d2ad5e6
commit 28e1299e60
21 changed files with 438 additions and 245 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/source/models/vlm.rst
View File

@@ -27,7 +27,7 @@ The :class:`~vllm.LLM` class can be instantiated in much the same way as languag
We have removed all vision language related CLI args in the ``0.5.1`` release. **This is a breaking change**, so please update your code to follow
the above snippet. Specifically, ``image_feature_size`` can no longer be specified as we now calculate that internally for each model.
To pass an image to the model, note the following in :class:`vllm.inputs.PromptInputs`:
To pass an image to the model, note the following in :class:`vllm.inputs.PromptType`:
* ``prompt``: The prompt should follow the format that is documented on HuggingFace.
* ``multi_modal_data``: This is a dictionary that follows the schema defined in :class:`vllm.multimodal.MultiModalDataDict`.