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

[Kernel][Core] Add AWQ support to the Marlin kernel (#6612)

Browse Source
This commit is contained in:
Alexander Matveev
2024-07-21 19:41:42 -04:00
committed by GitHub
parent 25e778aa16
commit 396d92d5e0
21 changed files with 1594 additions and 276 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/quantization/test_configs.py
View File

@@ -44,9 +44,9 @@ MODEL_ARG_EXPTYPES = [
("LnL-AI/TinyLlama-1.1B-Chat-v1.0-GPTQ-4bit", "awq", "ERROR"),
# AUTOAWQ
("TheBloke/OpenHermes-2.5-Mistral-7B-AWQ", None, "awq"),
("TheBloke/OpenHermes-2.5-Mistral-7B-AWQ", None, "awq_marlin"),
("TheBloke/OpenHermes-2.5-Mistral-7B-AWQ", "awq", "awq"),
("TheBloke/OpenHermes-2.5-Mistral-7B-AWQ", "marlin", "ERROR"),
("TheBloke/OpenHermes-2.5-Mistral-7B-AWQ", "marlin", "awq_marlin"),
("TheBloke/OpenHermes-2.5-Mistral-7B-AWQ", "gptq", "ERROR"),
]