chenchenghao/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

enable auto-round quantization model (#6226)

Browse Source 
Signed-off-by: Zhang, Weiwei1 <weiwei1.zhang@intel.com>
This commit is contained in:
Weiwei
2025-09-08 13:05:35 +08:00
committed by GitHub
parent b67c277f86
commit c8295d2353
8 changed files with 528 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
python/sglang/test/test_utils.py
View File

@@ -87,6 +87,11 @@ DEFAULT_DEEPPEP_MODEL_NAME_FOR_TEST = "deepseek-ai/DeepSeek-V3-0324"
DEFAULT_AWQ_MOE_MODEL_NAME_FOR_TEST = (
"hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4"
)
DEFAULT_AUTOROUND_MODEL_NAME_FOR_TEST = (
"OPEA/Llama-3.2-11B-Vision-Instruct-qvision-int4-sym-inc", ## mllm auto_round:auto_gptq
"OPEA/Qwen2.5-0.5B-Instruct-int4-sym-inc", ## auto_round:auto_gptq
"Intel/Qwen2-0.5B-Instruct-int4-sym-AutoRound", ## auto_round:auto_awq
)
DEFAULT_ENABLE_THINKING_MODEL_NAME_FOR_TEST = "Qwen/Qwen3-30B-A3B"
DEFAULT_DEEPSEEK_W4AFP8_MODEL_FOR_TEST = "Barrrrry/DeepSeek-R1-W4AFP8"