meadow-kun/llm-speed bench

Run llm-speed bench in CI, upload a signed run, and post the result as a PR comment.

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
modelModel identifier passed to the chosen backend (e.g. mlx-community/Qwen2.5-7B-Instruct-4bit, qwen/qwen3-coder-next).yes
backendOne of: ollama, llama.cpp, vllm, mlx, exllamav2, hosted-api.yes
api-baseBase URL for the llm-speed API.nohttps://api.llm-speed.com
api-keyOptional submitter API key. Leave empty for anonymous uploads.no""
anonSubmit anonymously (no signing key). Default false.nofalse
workloadWorkload preset to run.noquick
comment-on-prPost the result back as a PR comment when running on a pull_request event.notrue
namedescription
run-idThe r_xxxx slug returned by the API.
decode-tpsDecode tokens-per-second (numeric).
run-urlFull https URL to the run page on llm-speed.com.
badge-urlFull https URL to the badge SVG.