xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-11 01:45:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Docs] Update the AWQ documentation to highlight performance issue (#1883)

Browse Source
This commit is contained in:
Simon Mo 2023-12-02 15:52:47 -08:00 committed by GitHub
parent f86bd6190a
commit 4cefa9b49b
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 6 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
docs/source/quantization/auto_awq.rst
View File

@ -3,6 +3,12 @@
AutoAWQ AutoAWQ
================== ==================
.. warning::
Please note that AWQ support in vLLM is under-optimized at the moment. We would recommend using the unquantized version of the model for better
accuracy and higher throughput. Currently, you can use AWQ as a way to reduce memory footprint. As of now, it is more suitable for low latency
inference with small number of concurrent requests. vLLM's AWQ implementation have lower throughput than unquantized version.
To create a new 4-bit quantized model, you can leverage `AutoAWQ <https://github.com/casper-hansen/AutoAWQ>`_. To create a new 4-bit quantized model, you can leverage `AutoAWQ <https://github.com/casper-hansen/AutoAWQ>`_.
Quantizing reduces the model's precision from FP16 to INT4 which effectively reduces the file size by ~70%. Quantizing reduces the model's precision from FP16 to INT4 which effectively reduces the file size by ~70%.
The main benefits are lower latency and memory usage. The main benefits are lower latency and memory usage.