---
title: "Get Task Models"
method: GET
path: "/optimisation/tasks/{task_id}/models"
tags: ["optimisation"]
---

# Get Task Models

`GET /optimisation/tasks/{task_id}/models`

## Request

- Base URL: `https://api.os3robotics.com` (the document declares no server; this is the origin it was published from)
- URL: `https://api.os3robotics.com/optimisation/tasks/{task_id}/models`
- Auth: one of:
  - HTTP bearer
  - HTTP bearer

The model sizes and training methods the task's specialist can train on in the workspace,
each with its training estimate on `examples` requests and its serving cost, and the one
chosen for the task (null: OS3's default).

## Path parameters

- `task_id` string, uuid, required

## Query parameters

- `workspace_id` string, uuid, required
- `examples` integer

## Response `200`

Successful Response

- TaskModels: The task's model choices and the one chosen for it (null: OS3's default, then the larger size if the default does not pass).
  - `choices` ModelChoice[], required
    - `base_model` string, required
    - `name` string, required
    - `params` integer, required
    - `method` 'lora', required
    - `default` boolean, required
    - `training` Quote, required
      - `seconds` string, required
      - `gpu_seconds` string, required
      - `sku` string, required
      - `h100_seconds` string, required
      - `h100_seconds_max` string, required
      - `gpu` string, required
      - `effective_batch` integer, required
      - `samples` integer, required
      - `episodes` integer
      - `hours` string, required
      - `gpu_hours` string, required
      - `cost_usd` string, required
      - `remaining_h100_seconds` string, required
      - `billable_usd` string, required
      - `upgrade` Quote (recursive)
      - `depth_streams` string[]
    - `refusal` string, nullable, required
    - `serving` EndpointQuote, required
      - `gpu` string, required
      - `sku` string, required
      - `factor` string, required
      - `h100_seconds_per_day` string, required
      - `h100_hours_per_day` string, required
      - `usd_per_day` string, required
      - `h100_hour_usd` string, required
      - `remaining_h100_seconds` string, required
      - `included_days` string, nullable, required
      - `gpu_share` string
      - `token_rates` Rates: USD per 1M tokens OS3 endpoints serve: input the engine did not read from its prefix cache, input it did (``cached_input_per_1m``), and output, reasoning included.
        - `input_per_1m` string, nullable, required
        - `cached_input_per_1m` string, nullable, required
        - `output_per_1m` string, nullable, required
        - `provisional` boolean
  - `chosen` string, nullable, required

## Other responses

- `422`: Validation Error

---

[API](https://skmtc.dev/os3robotics/apis/os3-api.md) - [All operations](https://skmtc.dev/os3robotics/apis/os3-api/llms.txt) - [OpenAPI document](https://skmtc.dev/os3robotics/apis/os3-api/revisions/af32cd4f14bb?raw)
