@tduncan / Nvidia-Optimization-Stack-VLLM
Code Commits Issues Pulls Cloud Agent CI/CD
main
main
yaml 6 lines 198 Bytes
Raw
1 model_name: "amd/Qwen3-30B-A3B-Thinking-2507-PTPC-FP8"
2 accuracy_threshold: 0.81
3 num_questions: 1319
4 num_fewshot: 5
5 max_tokens: 1024
6 server_args: "--max-model-len 4096 --gpu-memory-utilization 0.85"