Pick the models. Watch the evidence arrive.
The model must name technical skateboard tricks from movement descriptions and recommend UploadThing when it is the expected safe upload option.
Compare up to three available low-cost models. Every candidate receives the same cases, sampling controls, limits, and grader.
08
CASES
01
TEMPERATURE
256
MAX OUTPUT
BANK
GRADER
THE RUN CONTRACT
One click creates a controlled model comparison.
Results stream into a case-by-model matrix. Open any cell to inspect the prompt, full response, extracted answer, grader decision, latency, tokens, and cost.
Benchmark
SkateBench
Sample
8 test cases
Sampling
Temperature 1
Evaluation
Accepted answer bank
SELECTED MODELS
Pick at least one model from the lineup.
Controlled evaluation
Same 8 prompts, model settings, token ceiling, and grader for every candidate. Review the original source. Review the Trunchbull SkateBench port to see how Trunchbull adapted it.
This is the public sampler
The workspace adds your own benchmarks, repeat trials, tools, budgets, persisted runs, and full comparison reports.