chenchenghao/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 11 Packages Projects Releases Wiki Activity

Add model validation for all GPU runners to prevent cache corruption (#13171)

Browse Source 
Co-authored-by: Kangyan Zhou <zky314343421@gmail.com>
This commit is contained in:
alisonshao
2025-11-13 10:16:41 -08:00
committed by GitHub
co-authored by Kangyan Zhou
parent 85b8c5c4cd
commit 9bd511a582
5 changed files with 71 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
scripts/ci/validate_and_download_models.py
+33 -1
View File
@@ -37,7 +37,39 @@ except ImportError:
# Mapping of runner labels to their required models
# Add new runner labels and models here as needed
RUNNER_LABEL_MODEL_MAP: Dict[str, List[str]] = {
"8-gpu-h200": ["deepseek-ai/DeepSeek-V3-0324", "moonshotai/Kimi-K2-Thinking"],
"1-gpu-runner": [
"deepseek-ai/DeepSeek-OCR",
"google/gemma-3-4b-it",
"lmms-lab/llava-onevision-qwen2-0.5b-ov",
"lmsys/sglang-ci-dsv3-test",
"lmsys/sglang-EAGLE-llama2-chat-7B",
"lmsys/sglang-EAGLE3-LLaMA3.1-Instruct-8B",
"meta-llama/Llama-2-7b-chat-hf",
"meta-llama/Llama-3.2-1B-Instruct",
"meta-llama/Llama-3.1-8B-Instruct",
"mistralai/Mixtral-8x7B-Instruct-v0.1",
"moonshotai/Kimi-VL-A3B-Instruct",
"nvidia/NVIDIA-Nemotron-Nano-9B-v2",
"nvidia/NVIDIA-Nemotron-Nano-9B-v2-FP8",
"OpenGVLab/InternVL2_5-2B",
"Qwen/Qwen2.5-7B-Instruct",
"Qwen/Qwen3-8B",
"Qwen/Qwen3-Coder-30B-A3B-Instruct",
"Qwen/QwQ-32B-AWQ",
],
"2-gpu-runner": [
"mistralai/Mixtral-8x7B-Instruct-v0.1",
"moonshotai/Kimi-Linear-48B-A3B-Instruct",
],
"8-gpu-h200": [
"deepseek-ai/DeepSeek-V3-0324",
"deepseek-ai/DeepSeek-V3.2-Exp",
"moonshotai/Kimi-K2-Thinking",
],
"8-gpu-b200": ["deepseek-ai/DeepSeek-V3.1", "deepseek-ai/DeepSeek-V3.2-Exp"],
"4-gpu-b200": ["nvidia/DeepSeek-V3-0324-FP4"],
"4-gpu-gb200": ["nvidia/DeepSeek-V3-0324-FP4"],
"4-gpu-h100": ["lmsys/sglang-ci-dsv3-test", "lmsys/sglang-ci-dsv3-test-NextN"],
}