---
title: "Get Base Model Benchmarks"
method: GET
path: "/base-models/{model_id}/benchmarks"
tags: ["model-selection"]
---

# Get Base Model Benchmarks

`GET /base-models/{model_id}/benchmarks`

Third-party benchmark figures and axis means for one base model.

Args:
    model_id: Canonical catalog model ID.
    auth: Optional caller identity; benchmarks are public reference data.

Returns:
    Roster benchmarks grouped by axis, plus the per-axis means.

Raises:
    HTTPException: 404 when the model is not in the catalog.

## Path parameters

- `model_id` string, required

## Response `200`

Successful Response

- ModelBenchmarksResponse — Roster benchmarks and axis means for a single base model.
  - `model_id` string, required — Canonical catalog model ID
  - `coding` BenchmarkScore[] — Coding benchmarks, strongest first
    - `benchmark_id` string, required — Upstream benchmark slug
    - `label` string, required — Display name for the benchmark
    - `axis` 'coding' | 'agentic', required — Which axis the benchmark measures
    - `score` number, required — Normalized score on a 0-100 scale
    - `self_reported` boolean, required — Whether the vendor reported the figure
    - `verified` boolean, required — Whether the source independently verified it
    - `caveat` string, nullable — Qualification to show beside the figure
    - `rank` integer, nullable — Position on this benchmark upstream
    - `rank_total` integer, nullable — How many models reported this benchmark upstream
    - `scored_at` string, date-time, nullable — When the figure was published
  - `agentic` BenchmarkScore[] — Agentic benchmarks, strongest first
    - `benchmark_id` string, required — Upstream benchmark slug
    - `label` string, required — Display name for the benchmark
    - `axis` 'coding' | 'agentic', required — Which axis the benchmark measures
    - `score` number, required — Normalized score on a 0-100 scale
    - `self_reported` boolean, required — Whether the vendor reported the figure
    - `verified` boolean, required — Whether the source independently verified it
    - `caveat` string, nullable — Qualification to show beside the figure
    - `rank` integer, nullable — Position on this benchmark upstream
    - `rank_total` integer, nullable — How many models reported this benchmark upstream
    - `scored_at` string, date-time, nullable — When the figure was published
  - `axes` BenchmarkAxisSummary[] — Per-axis summaries
    - `axis` 'coding' | 'agentic', required — Which axis the mean summarises
    - `mean` number, nullable — Mean of the benchmarks, 0-100
    - `count` integer, required — Benchmarks behind the mean
  - `source_note` string, nullable — Attribution for the underlying figures
  - `updated_at` string, date-time, nullable — Most recent time we refreshed these figures

## Other responses

- `422` — Validation Error

---

[API](https://skmtc.net/pioneer/apis/brain-api.md) · [All operations](https://skmtc.net/pioneer/apis/brain-api/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/pioneer/brain-api/versions/31dfe831e079/schema)