# Benchmark

URL: https://docs.hanzo.ai/docs/cli/benchmark

> The `hanzo benchmark` commands.

Also for this capability: [API](https://docs.hanzo.ai/docs/openapi/benchmark) · **CLI** · [MCP](https://docs.hanzo.ai/docs/mcp-tools/benchmark) · [SDKs](https://docs.hanzo.ai/docs/sdks)

One honest score for any model, on the tests everyone quotes.

[API reference →](https://docs.hanzo.ai/docs/openapi/benchmark)

```bash
hanzo benchmark catalog
```

Every command takes `--json` for the raw response and `--help` for its own flags. Sign in once with `hanzo auth login`; the commands below use that session, and the org they act in is the one it carries.

| Command                                                | What it does                                                                                                                                                                                                                  |
| ------------------------------------------------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `hanzo benchmark catalog`                              | The canonical public benchmarks this arena runs — the id, title, axis, item count and upstream source of each, with native marking the ones the standardized harness runs today; the rest are registered and adapter-pending. |
| `hanzo benchmark compare --a <a> --b <b>`              | The ONLY valid arm-vs-arm test: it pairs the two models on the items BOTH completed, and answers rescue and damage counts with an exact-McNemar p.                                                                            |
| `hanzo benchmark history`                              | Returns each model's measured score per run over time, oldest first, with the change between runs.                                                                                                                            |
| `hanzo benchmark leaderboard`                          | Answers one row per model for the benchmark named — what our own harness measured, beside what the vendor claims, and the gap between them.                                                                                   |
| `hanzo benchmark runs --benchmarks '["<benchmarks>"]'` | Admits and queues a benchmark run against a model or your own endpoint, and answers 202 with the receipt.                                                                                                                     |

### Claims

| Command                         | What it does                                                                                                          |
| ------------------------------- | --------------------------------------------------------------------------------------------------------------------- |
| `hanzo benchmark claims create` | Records claims for the caller's org: one to correct a number, many to import a leaderboard.                           |
| `hanzo benchmark claims get`    | Lists the effective claims the caller may read, each labelled with the org that made it and the user who recorded it. |

### Presets

| Command                          | What it does                                                                                                                                                                                          |
| -------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `hanzo benchmark presets create` | Validates a router blend — its name, its arms, the rank they escalate through and the panel fan-out width — and answers 202 with the preset and the enso-\<name> it would be served as.               |
| `hanzo benchmark presets get`    | Are the router blends available to compose from — a named set of model arms, the rank they escalate through and the panel width that bounds fan-out — each served by the model layer as enso-\<name>. |