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

[Kernel/Quant] Remove the original marlin format and qqq (#23204)

Browse Source 
Signed-off-by: mgoin <mgoin64@gmail.com>
This commit is contained in:
Michael Goin
2025-08-20 15:13:36 -04:00
committed by GitHub
parent ebe56a0064
commit 0cdbf5e61c
26 changed files with 92 additions and 3698 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/weight_loading/models.txt
View File

@@ -26,9 +26,5 @@ compressed-tensors, nm-testing/SparseLlama-3.1-8B-gsm8k-pruned.2of4-W8A8-testing
awq, casperhansen/mixtral-instruct-awq, main
awq_marlin, casperhansen/mixtral-instruct-awq, main
fp8, neuralmagic/Meta-Llama-3-8B-Instruct-FP8-KV, main
marlin, nm-testing/zephyr-beta-7b-marlin-g128, main
marlin, robertgshaw2/zephyr-7b-beta-channelwise-marlin, main
qqq, HandH1998/QQQ-Llama-3-8b-g128, main
qqq, HandH1998/QQQ-Llama-3-8b, main
hqq, nm-testing/Llama-3.2-1B-Instruct-HQQ, main
None, mgleize/fairseq2-dummy-Llama-3.2-1B, main