Model

A model the project can call.

{
  "id": "openai/gpt-oss-120b",
  "name": "GPT OSS 120B",
  "ownedBy": "openai",
  "createdAt": "2026-09-30T00:00:00.000Z",
  "status": "available",
  "inputs": ["text"],
  "contextWindow": 128000,
  "maxOutputTokens": 16000,
  "defaultOutputTokens": 4096,
  "tools": true,
  "reasoning": true,
  "units": { "input": 20000, "cached": 20000, "output": 79000 },
  "processedIn": "Germany",
  "retiresAt": null,
  "replacedBy": null
}

Fields

idstring

The model’s name, to pass as model to chat().

namestring

The model’s name to show people, such as GPT OSS 120B.

ownedBystring

Who makes the model, such as openai or nvidia.

createdAtDate

When the model was added.

statusstring

preview, available, or deprecated. A preview model can change or be removed without notice.

inputsstring[]

What the model reads: text, and image for a model that reads images.

contextWindownumber

The most tokens the messages and the answer can hold together.

maxOutputTokensnumber

The highest maxTokens the model takes.

defaultOutputTokensnumber

The answer’s limit when you leave out maxTokens.

toolsboolean

Whether the model can call tools.

reasoningboolean

Whether the model thinks before it answers, and so takes reasoningEffort.

unitsobject

The AI units per million tokens: input read, cached read from the cache, and output written. See AI units.

processedInstring

The country where the model runs.

retiresAtDate or null

For a deprecated model, the day it stops answering.

replacedBystring or null

For a deprecated model, the model to move to.

Returned by

• space.ai.models()