Trillet AI logo

List Available Models

List the language models currently available to your workspace.

GEThttps://api.trillet.ai/v1/api/llm-models/call-flow-models

Returns every language model your workspace can use, so you can pass a valid id as llmModel when creating or updating a call agent.

The available models change over time. Read this endpoint rather than hardcoding a list.

To list the models available to one particular call flow, call GET /v1/api/llm-models/call-flow-models/{callFlowId}. It returns the same fields, and also includes models enabled only for that flow.

Headers

x-workspace-idstringrequired

The unique identifier of the workspace being accessed.

Response Fields

idstring

The model identifier. This is the value to pass as llmModel on agent create and update.

namestring

Display name for the model.

descriptionstring

Short description, usually the context window. May be empty.

speedstring

Relative speed score, as a percentage. May be empty.

qualitystring

Relative quality score, as a percentage. May be empty.

tierstring

Broad performance band, for example fast or balanced.

providerstring

The model's vendor.

tagsstring[]

Labels shown in the model picker. May be empty.

standardRatePerMinutenumber

Your workspace's standard per-minute call rate.

effectiveRatePerMinutenumber

What a minute costs your workspace with this model, including any model surcharge.

costMultipliernumber | null

How many times the standard rate this model costs. null when it costs no more than a standard minute.

isRecommendedboolean

Whether the model is recommended for general use.

isTopPickboolean

Whether the model is the current default choice. Exactly one model carries this.

averageLatencynumber

Mean time to first token in seconds, measured across recent calls. Present only once a model has enough traffic to report on.

Response

[
  {
    "id": "platform-llm",
    "name": "Platform LLM",
    "description": "",
    "speed": "",
    "quality": "",
    "tier": "fast",
    "tags": [],
    "isRecommended": true,
    "isTopPick": true,
    "allowedFlowIds": [],
    "averageLatency": 0.32
  },
  {
    "id": "gpt-4.1",
    "name": "GPT-4.1",
    "description": "1M context window",
    "speed": "93%",
    "quality": "94%",
    "tier": "balanced",
    "tags": [],
    "isRecommended": true,
    "isTopPick": false,
    "allowedFlowIds": [],
    "averageLatency": 0.88
  },
  {
    "id": "gemini-2.5-flash",
    "name": "Gemini 2.5 Flash",
    "description": "1M context window",
    "speed": "99%",
    "quality": "98%",
    "tier": "balanced",
    "tags": [],
    "isRecommended": false,
    "isTopPick": false,
    "allowedFlowIds": [],
    "averageLatency": 0.83
  }
]

Last updated on