Hanzo
OpenapiBenchmark

Admits and queues a benchmark run against a model or your own endpoint, and…

Admits and queues a benchmark run against a model or your own endpoint, and answers 202 with the receipt.

POST /v1/benchmark/runs

Addresshttps://api.hanzo.ai/v1/benchmark/runs
MethodPOST
Operationpost_benchmark_runs
AuthAuthorization: Bearer $HANZO_API_KEY

Admits and queues a benchmark run against a model or your own endpoint, and answers 202 with the receipt.

It is an ADMISSION, not a result: the work is done by the harness afterwards and the numbers appear on the leaderboard as it completes them.

Cost is bounded by the store rather than by a quota: attempts are append-only and keyed by (benchmark, item, model), so an (item, model) pair already attempted is skipped instead of re-spent, and re-queuing the same run is close to free.

Validation is up front and total — a request with neither model nor endpoint is a 400, one with no benchmarks is a 400, and any benchmark id outside the catalog is a 422 naming exactly which ids were unknown, so a typo never silently queues a partial run.

Request

4 fields, body application/json (required).

FieldInTypeRequiredDescription
attemptsbodyintegerAttempts is how many times to try each item; the harness's default applies when it is omitted.
benchmarksbodystring[]yesBenchmarks are the catalog ids to run.
endpointbodystringEndpoint is your own chat-completions URL, for benchmarking a model this arena does not host.
modelbodystringModel is the catalog model id to run.

Response

StatusBodyMeaning
202admissionaccepted

202 body — 5 fields.

FieldInTypeAlwaysDescription
benchmarksbodystring[]Benchmarks are the catalog ids admitted.
endpointbodystringEndpoint is the caller's own endpoint the run targets.
modelbodystringModel is the catalog model the run targets.
notebodystringNote explains what admission does and does not promise.
statusbodystringStatus is "queued": the run is admitted, not finished.

Failure carries the platform error shape — see Errors.

Examples

hanzo benchmark runs --benchmarks '["<benchmarks>"]'

Benchmark API · All Hanzo APIs · Interactive reference

How is this guide?

On this page