---
title: "Get recommended base models"
method: GET
path: "/v2/workspaces/{workspace_id}/base-models/defaults"
tags: ["Base Models"]
---

# Get recommended base models

`GET /v2/workspaces/{workspace_id}/base-models/defaults`

Layer's curated recommendations, filtered to the models this workspace may run. Use this to pick a model when the caller named a modality or an intent but not a specific `base_model_id`: take entry `[0]` of the matching bucket, falling through to later entries if it is unavailable. Prefer `preferred_by_use_case` over `featured_by_modality` when the intent is known, since it is finer-grained.

The curation changes as models ship and are retired, so clients should re-read this periodically rather than hard-coding a model id.

## Path parameters

- `workspace_id` string, uuid, required — Id of the workspace that owns the resource.

## Response `200`

Successful Response

- BaseModelDefaultsOutput — Layer's curated recommendations, filtered to what this workspace may run. Both maps are ordered best-first, and entry ``[0]`` is the pick to use when the caller expresses no preference. Later entries are genuine fallbacks, so a client holding a stale copy can walk the list. Buckets left empty by workspace filtering are omitted rather than returned empty, so a present key always has a usable pick.
  - `featured_by_modality` object, required — Recommended base models per output modality, best first. `[0]` is the default for a request that names only a modality.
  - `preferred_by_use_case` object, required — Recommended base models per use case, best first. Finer-grained than modality — prefer this when the caller's intent is known.
  - `base_models` BaseModelSummary[], required — Details for every base model named above, deduplicated, so a client can render a recommendation without a second request.
    - `base_model_id` string, required — Base model identifier, e.g. flux-dev — pass it as `base_model_id`.
    - `name` string, required — Display name.
    - `aliases` string[] — Community nicknames this model is also known by.
    - `modality` 'text' | 'image' | 'audio' | 'video' | 'three_d' | 'playable', required
    - `description` string, nullable — Short description.
    - `price_per_unit` number, nullable — CU cost per measurement unit.
    - `price_unit` string, nullable — Measurement unit, e.g. megapixel, second, generation.
    - `inference_timing` string, required — Typical generation time range, e.g. '5-15s'.

## Other responses

- `401` — Unauthenticated — missing or invalid Bearer token.
- `403` — Forbidden — insufficient permissions, or the access token lacks the scope the operation requires.
- `404` — Resource not found.
- `409` — The request with this `Idempotency-Key` is still running — retry after the number of seconds in `Retry-After`.
- `422` — Invalid input parameters, or an `Idempotency-Key` reused for a different request.
- `429` — Rate limited — retry after the number of seconds in `Retry-After`.
- `500` — Internal server error.

---

[API](https://skmtc.dev/layer/apis/layer-rest-api.md) · [All operations](https://skmtc.dev/layer/apis/layer-rest-api/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/layer/layer-rest-api/revisions/fafc779f9ba2/schema)