Hanzo
CLIBenchmark

Benchmark

9 `hanzo` commands for Benchmark, each calling one operation on api.hanzo.ai.

Package benchmark is one honest score for any model, on the tests everyone quotes.

9 commands · API reference →

CommandCallsWhat it does
hanzo benchmark catalogGET /v1/benchmark/catalogIs the canonical public benchmarks this arena runs — the id, title, axis, item count and upstream source of each, with…
hanzo benchmark claims getGET /v1/benchmark/claimsLists the effective published claims: what the leaderboard will use for each (benchmark, model) after the seed, the…
hanzo benchmark claims createPOST /v1/benchmark/claimsRecords published claims: one to correct a number, many to import a leaderboard.
hanzo benchmark compare --a <a> --b <b>GET /v1/benchmark/compareIs the ONLY valid arm-vs-arm test: it pairs the two models on the items BOTH completed, and answers rescue and damage…
hanzo benchmark historyGET /v1/benchmark/historyReturns each model's measured score per run over time, oldest first, with the change between runs.
hanzo benchmark leaderboardGET /v1/benchmark/leaderboardAnswers one row per model for the benchmark named — what our own harness measured, beside what the vendor claims, and…
hanzo benchmark presets getGET /v1/benchmark/presetsAre the router blends available to compose from — a named set of model arms, the rank they escalate through and the…
hanzo benchmark presets createPOST /v1/benchmark/presetsValidates a router blend — its name, its arms, the rank they escalate through and the panel fan-out width — and answers…
hanzo benchmark runs --benchmarks '["<benchmarks>"]'POST /v1/benchmark/runsAdmits and queues a benchmark run against a model or your own endpoint, and answers 202 with the receipt.

How is this guide?