Skip to main content
GET/compute/llms/{modelId}

Get LLM Model

Retrieves metadata for a single LLM model by its identifier, including name, provider class, modality, token cost (per million input and output tokens), and supported parameters.

Requires any Forge, Studio, or Workspace cohort role.

Parameters
modelId *path

The model's id, as the model list returns it.

projectSlug *path

The project's slug: the path segment after @.

It is unique within the org.

Responses
HTTPK9O CodeDescription
200Success
400K9O_MODEL_NOT_DECLAREDThe requested model ID is not in the supported model catalogue.
401K9O_INVALID_CREDENTIALSThe request credentials are missing or invalid.
401K9O_SESSION_EXPIREDYour session has expired; sign in again to continue.
403K9O_ACCESS_DENIEDYour role does not grant access to read model metadata for this project.
Response body (200)
data *object

A model in the project's model library.

id *string

The model's id.

Pass it as modelId to read or invoke the model.

provider *"self_hosted" | "anthropic" | "openai"

Who serves the model. self_hosted: the platform runs it. anthropic or openai: the outside provider, called with the project's own key.

providerClass *"platform" | "tenant_key"

platform: run and billed by the platform. tenant_key: called with the project's own provider key and billed by that provider.

modality *"text_completion" | "text_completion_with_tools" | "embeddings"

What the model does. text_completion: generates text. text_completion_with_tools: generates text and can call tools. embeddings: turns text into vectors.

invocable *boolean

Whether this model can be invoked through the platform.

reason"tenant_key_only"

Present only when invocable is false.

priceIn *any

US dollars per 1M input tokens (embedding tokens for an embeddings model).

Null for a tenant_key model or when no rate is set.

One of: number
or: null
priceOut *any

US dollars per 1M output tokens.

Null for an embeddings model, a tenant_key model, or when no rate is set.

One of: number
or: null
priceUnit *"usd_per_million_tokens"

The unit of priceIn and priceOut: always US dollars per million tokens.

meta *any
TypeScript
// const k9o = new K9o({ orgSlug: 'playground', projectSlug: 'storefront', apiKey: 'k9o_…' })
const { data } = await k9o.compute.llms.read('mistral-7b')

Showing example in TypeScript

Response example (200)
{
  "data": {
    "id": "string",
    "provider": "self_hosted",
    "providerClass": "platform",
    "modality": "text_completion",
    "invocable": false,
    "reason": "tenant_key_only",
    "priceIn": 0,
    "priceOut": 0,
    "priceUnit": "usd_per_million_tokens"
  }
}

More in Compute & Automation