@tduncan / Nvidia-Optimization-Stack-VLLM
Code Commits Issues Pulls Cloud Agent CI/CD
main
main
yaml 8 lines 278 Bytes
Raw
1 model_name: "google/gemma-4-E4B-it-qat-mobile-ct"
2 accuracy_threshold: 0.50
3 num_questions: 1319
4 num_fewshot: 5
5 server_args: >-
6 --enforce-eager
7 --max-model-len 4096
8 --speculative-config '{"method":"mtp","model":"google/gemma-4-E4B-it-assistant","num_speculative_tokens":4}'