Models

GET /v1/models is the source of truth — call it before you hardcode an id anywhere, because the table below is a snapshot of that same call and this page can go stale between deploys the way any documentation can.

GET/v1/models

Response

{
  "object": "list",
  "data": [
    {"id": "data-auto", "object": "model", "owned_by": "openai"},
    {"id": "data-general", "object": "model", "owned_by": "openai"},
    {"id": "data-law-ir", "object": "model", "owned_by": "openai"},
    {"id": "data-law-es", "object": "model", "owned_by": "openai"},
    {"id": "data-stt", "object": "model", "owned_by": "openai"},
    {"id": "data-tts", "object": "model", "owned_by": "openai"},
    {"id": "data-embed", "object": "model", "owned_by": "openai"}
  ]
}

owned_by: "openai" is a compatibility artifact of the OpenAI wire format, not a claim about who trained the weights — every id above runs on this platform's own hardware. See Introduction.

Deprecated aliases

The previous model ids — every larsa-* and lardad-* name, for example data-general or data-law-ir — still route to the same models for now, but they are deprecated aliases and no longer listed by GET /v1/models. They will be removed after a notice period, so write new code against the data-* ids above and migrate existing code when convenient.

Reference

Model	For	Context	Modalities	Tools
`data-auto`	One id for everything — classifies your message from its text and content and dispatches to the right specialist below. See Chat.	262,144	text, image → text	Yes
`data-general`	The default chat and vision model — llama.cpp serving Qwen3.6-35B-A3B. Reads images and documents. See Chat and Vision.	262,144	text, image → text	Yes
`data-law-ir`	Iranian law, retrieval-grounded over 117,499 provisions before the model answers.	262,144	text → text	No
`data-law-es`	Spanish and EU law (BOE consolidated), retrieval-grounded over 835,951 provisions.	262,144	text → text	No
`data-stt`	Speech to text — Whisper large-v3, silence-based chunking for long audio. See Speech to text and Realtime transcription.	—	audio → text	—
`data-tts`	Text to speech — Aava for Persian, Chatterbox for 23 other languages, with a round-trip check against `data-stt` available on request. See Text to speech.	—	text → audio	—
`data-embed`	Embeddings — BAAI's `bge-m3`, 1024 dimensions, multilingual. The same server the retrieval corpora above are indexed with. See Embeddings.	8,192	text → vector	—

Two ceilings that apply to all of them

Context, 262,144 tokens. Both chat backends were started with the same window, and the retrieval models generate on the same weights after their search step — so it is one number across the board, not a per-model detail to look up.
Output, 8,192 tokens per request, platform-wide. A generation with no ceiling can run to the context limit and hold a GPU slot for everyone else — max_tokens may ask for less than 8,192, never more. Requesting more is clamped, not refused.

The public price list at /pricing also carries a handful of rows — gpt-4o, unknown, the wildcard * — used internally for cost estimation. They are not callable model ids; GET /v1/models is the list that matters.

Reference#

Two ceilings that apply to all of them#

Reference

Two ceilings that apply to all of them