TiinyBench
Measure what your Tiiny does under real work.
Install
Nothing opened? Get the launcher.
Then open http://localhost:8425. New here? Install the farm CLI first.
What it does
Times prefill against prompt length, throughput over a long unbroken generation, what happens when several callers arrive at once, and what a reasoning model charges in wall time for the tokens nobody reads. Image, speech and embedding models get their own measurements: seconds per 512 plate, real-time factor, embeddings per second. Results are plain JSON and the report is one self-contained HTML file with no CDN, no webfont and no JavaScript needed to read it. By default it loads and unloads nothing and benchmarks whatever is already running, so it is safe on a box doing real work. Python standard library only.
Screenshots
No screenshots yet.
Art in the farm's hand
Describe this app's scene in one sentence and the farm draws a header and an icon in the same style as every other app on the shelf. How this works.
Use these opens the update form with the new pair attached. A published app's images change through the same pull request as everything else.