40 lines
863 B
ReStructuredText
40 lines
863 B
ReStructuredText
.. _input_processing:
|
|
|
|
Input Processing
|
|
================
|
|
|
|
.. currentmodule:: vllm.inputs
|
|
|
|
Each model can override parts of vLLM's :ref:`input processing pipeline <input_processing_pipeline>` via
|
|
:data:`~vllm.inputs.INPUT_REGISTRY` and :data:`~vllm.multimodal.MULTIMODAL_REGISTRY`.
|
|
|
|
Currently, this mechanism is only utilized in :ref:`multi-modal <multi_modality>` models for preprocessing multi-modal input
|
|
data in addition to input prompt, but it can be extended to text-only language models when needed.
|
|
|
|
Guides
|
|
++++++
|
|
|
|
.. toctree::
|
|
:maxdepth: 1
|
|
|
|
input_processing_pipeline
|
|
|
|
Module Contents
|
|
+++++++++++++++
|
|
|
|
LLM Engine Inputs
|
|
-----------------
|
|
|
|
.. autoclass:: vllm.inputs.LLMInputs
|
|
:members:
|
|
:show-inheritance:
|
|
|
|
Registry
|
|
--------
|
|
|
|
.. autodata:: vllm.inputs.INPUT_REGISTRY
|
|
|
|
.. automodule:: vllm.inputs.registry
|
|
:members:
|
|
:show-inheritance:
|