---
title: "Get Llms Presets"
method: GET
path: "/v1/models/llms/presets"
tags: ["V1 - Platform"]
---

# Get Llms Presets

`GET /v1/models/llms/presets`

## Request

- Base URL: `https://api.app.noxus.ai` (the document declares no server; this is the origin it was published from)
- URL: `https://api.app.noxus.ai/v1/models/llms/presets`
- Auth: API key in header `X-API-Key`

## Response `200`

Successful Response

- SpotflowModelPreset[]
  - `handle` string, nullable: Unique handle for the preset, e.g. 'cost', 'chat-quality'
  - `icon` string, nullable: Icon identifier for the preset, e.g. 'dollar', 'gauge', 'zap'
  - `models` union[], required: List of model IDs
    - union
      - string
      - ModelPreset
        - `handle` string, nullable: Unique handle for the preset, e.g. 'cost', 'chat-quality'
        - `icon` string, nullable: Icon identifier for the preset, e.g. 'dollar', 'gauge', 'zap'
        - `models` union[], required: List of model IDs
          - union
            - string
            - ModelPreset (recursive)
        - `required_capabilities` ModelCapability[]: Capabilities required for the preset
        - `capability_check_mode` 'any' | 'all': Whether a model must have 'all' required capabilities or just 'any' one of them
        - `minimum_speed_tier` 'low' | 'medium' | 'high'
        - `minimum_quality_tier` 'low' | 'medium' | 'high'
        - `title` string: Title of the preset
        - `subtitle` string: Subtitle of the preset
        - `description` string: Description of the preset
        - `parameters` ModelParameters
          - `temperature` union: Controls randomness in generation
            - number
            - 'NOT_GIVEN'
          - `top_p` union: Nucleus sampling threshold
            - number
            - 'NOT_GIVEN'
          - `top_k` union: Top-k sampling threshold
            - integer
            - 'NOT_GIVEN'
          - `frequency_penalty` union: Penalty for token frequency
            - number
            - 'NOT_GIVEN'
          - `presence_penalty` union: Penalty for token presence
            - number
            - 'NOT_GIVEN'
          - `logit_bias` union: Token biasing dictionary
            - object
            - 'NOT_GIVEN'
          - `max_tokens` union: Maximum number of tokens to generate
            - integer
            - 'NOT_GIVEN'
          - `stop` union: Stop sequences
            - string[]
            - 'NOT_GIVEN'
          - `logprobs` union: Number of logprobs to return
            - integer
            - 'NOT_GIVEN'
          - `top_logprobs` union: Number of most likely tokens to return
            - integer
            - 'NOT_GIVEN'
          - `seed` union: Random seed for reproducibility
            - integer
            - 'NOT_GIVEN'
          - `timeout` union: Request timeout in seconds
            - number
            - 'NOT_GIVEN'
          - `include_reasoning` union: Whether to include reasoning steps
            - boolean
            - 'NOT_GIVEN'
          - `reasoning_effort` union: Reasoning level
            - 'off' | 'low' | 'medium' | 'high' | 'xhigh'
            - 'NOT_GIVEN'
          - `parallel_tool_calls` union: Whether to allow parallel function calling
            - boolean
            - 'NOT_GIVEN'
          - `model_override` union: Model override parameters
            - object
            - 'NOT_GIVEN'
        - `data_center_fallback` boolean: Whether to use data center fallback, meaning to change regions of the model when available
        - `exhaustive` boolean: Whether to exhaustively search for models in the registry, after preset if done
  - `required_capabilities` ModelCapability[]: Capabilities required for the preset
  - `capability_check_mode` 'any' | 'all': Whether a model must have 'all' required capabilities or just 'any' one of them
  - `minimum_speed_tier` 'low' | 'medium' | 'high'
  - `minimum_quality_tier` 'low' | 'medium' | 'high'
  - `title` string: Title of the preset
  - `subtitle` string: Subtitle of the preset
  - `description` string: Description of the preset
  - `parameters` ModelParameters
    - `temperature` union: Controls randomness in generation
      - number
      - 'NOT_GIVEN'
    - `top_p` union: Nucleus sampling threshold
      - number
      - 'NOT_GIVEN'
    - `top_k` union: Top-k sampling threshold
      - integer
      - 'NOT_GIVEN'
    - `frequency_penalty` union: Penalty for token frequency
      - number
      - 'NOT_GIVEN'
    - `presence_penalty` union: Penalty for token presence
      - number
      - 'NOT_GIVEN'
    - `logit_bias` union: Token biasing dictionary
      - object
      - 'NOT_GIVEN'
    - `max_tokens` union: Maximum number of tokens to generate
      - integer
      - 'NOT_GIVEN'
    - `stop` union: Stop sequences
      - string[]
      - 'NOT_GIVEN'
    - `logprobs` union: Number of logprobs to return
      - integer
      - 'NOT_GIVEN'
    - `top_logprobs` union: Number of most likely tokens to return
      - integer
      - 'NOT_GIVEN'
    - `seed` union: Random seed for reproducibility
      - integer
      - 'NOT_GIVEN'
    - `timeout` union: Request timeout in seconds
      - number
      - 'NOT_GIVEN'
    - `include_reasoning` union: Whether to include reasoning steps
      - boolean
      - 'NOT_GIVEN'
    - `reasoning_effort` union: Reasoning level
      - 'off' | 'low' | 'medium' | 'high' | 'xhigh'
      - 'NOT_GIVEN'
    - `parallel_tool_calls` union: Whether to allow parallel function calling
      - boolean
      - 'NOT_GIVEN'
    - `model_override` union: Model override parameters
      - object
      - 'NOT_GIVEN'
  - `data_center_fallback` boolean: Whether to use data center fallback, meaning to change regions of the model when available
  - `exhaustive` boolean: Whether to exhaustively search for models in the registry, after preset if done
  - `type` 'spotflow'

---

[API](https://skmtc.dev/noxus/apis/noxus-backend.md) - [All operations](https://skmtc.dev/noxus/apis/noxus-backend/llms.txt) - [OpenAPI document](https://skmtc.dev/noxus/apis/noxus-backend/revisions/0e37810b7ad5?raw)