@tduncan / Nvidia-Optimization-Stack-VLLM
Code Commits Issues Pulls Cloud Agent CI/CD
main
main
yaml 9 lines 231 Bytes
Raw
1 model_name: "RedHatAI/Qwen3.5-4B-quantized.w4a16"
2 accuracy_threshold: 0.82
3 num_questions: 1319
4 num_fewshot: 5
5 gen_prefix: " <think>\n\n</think>\n"
6 server_args: >-
7 --enforce-eager
8 --max-model-len 8192
9 --linear-backend humming