Profile:
Model:
Profile:
Model AI Throughput Evaluator — reproducible LLM inference benchmarks. mate-bench on GitHub
Run mate run llm --profile standard and submit.
| # | GPU | VRAM | Profile | Model | Tokens/s | Runtime | Runs | Last submitted |
|---|
Run mate run llm --mode open --model your-model --profile standard and submit.
| # | GPU | VRAM | Model | Profile | Tokens/s | Runtime | Runs | Last submitted |
|---|