Core
List Models
List models currently available on TokenLab
The model catalog is public and does not require an API key. Each item includes a summary of current capabilities, prices, and API formats. Retrieve one model when you need full details.
When you provide an API key, TokenLab validates it and filters the catalog for its model permissions and effective delivery policy. A key set to verified lists models with TokenLab Verified supply; official lists models with Official supply. Refresh your model selector when the key or delivery policy changes. The request's supported delivery override takes precedence over the key and workspace defaults.
Query Parameters
Filter by chat, image, video, audio, tts, stt, music, 3d, embedding, rerank, decision, or translation.
Return current recommendations for a task such as image, video, music, 3d, tts, stt, embedding, rerank, decision, or translation.
Filter by model maker, such as openai, anthropic, google, or deepseek.
Filter by a capability tag such as reasoning, tool-use, vision, or text-to-image. The values in a model's tokenlab.capabilities are valid tags. To filter by kind of model, use category.
fullfull returns the complete list shape described below. compact returns a smaller list for model selection: id, provider, category, capabilities, token limits, a price summary, status, deliveryAvailability, supported_operations, and request_endpoint, plus view and count at the top level.
Response
list.id, owned_by, and a tokenlab summary.Common list fields under tokenlab include:
tokenlab.aliasestokenlab.capabilitiestokenlab.deliveryAvailability: configuredverifiedandofficialsupport, when provided. A missing field means unknown; support does not guarantee real-time availability.tokenlab.pricing: the Verified price. Its price fields arenullwhentokenlab.deliveryAvailability.verifiedisfalse; such a model is billed at the Official price.tokenlab.official_pricing: the Official price. Requests that Official delivers are billed at this price.tokenlab.pricing_unittokenlab.max_input_tokensandtokenlab.max_output_tokenstokenlab.categorytokenlab.lifecycle,tokenlab.commercial, andtokenlab.badgestokenlab.request_format_summarytokenlab.agent_preferenceswhenrecommended_foris used
When recommended_for is used, results may also include recommendation status and an update time. Read the current response instead of pinning the order.
Errors
| Status | Code | Cause |
|---|---|---|
400 | invalid_view, invalid_provider, invalid_category, invalid_recommended_for | A query value is not accepted. The message lists the valid values. category and recommended_for must match when both are set |
401 | invalid_api_key | You sent an API key that is not valid. Remove the header to read the public list |
429 | rate_limit_exceeded | Too many requests to the public catalog. Cache the response and retry after Retry-After |
Example
curl "https://api.tokenlab.sh/v1/models?category=video"{
"object": "list",
"data": [
{
"id": "veo3.1",
"object": "model",
"owned_by": "google",
"tokenlab": {
"category": "video",
"capabilities": ["text-to-video", "image-to-video"]
}
}
]
}See Get a Model for complete details.
curl -X GET "https://example.com/v1/models"{ "object": "list", "data": [ { "id": "string", "object": "model", "created": 0, "owned_by": "string", "tokenlab": { "aliases": [ "string" ], "deliveryAvailability": { "verified": true, "official": true, "operations": { "property1": { "verified": true, "official": true }, "property2": { "verified": true, "official": true } } }, "pricing": { "input_per_1m": "string", "output_per_1m": "string", "per_request": "string", "is_lock_price": true, "currency": "USD" }, "official_pricing": { "input_per_1m": "string", "output_per_1m": "string", "per_request": "string", "is_lock_price": true, "currency": "USD" }, "capabilities": [ "string" ], "providers": [ "string" ], "max_input_tokens": 0, "max_output_tokens": 0, "category": "string", "pricing_unit": "string", "cache_pricing": {}, "official_cache_pricing": {}, "has_complex_pricing": true, "pricing_summary": {}, "agent_preferences": {}, "lifecycle": { "stage": "active", "reason": "string", "replacement_model": "string", "released_at": "2019-08-24T14:15:22Z", "deprecated_at": "2019-08-24T14:15:22Z", "retired_at": "2019-08-24T14:15:22Z", "sunset_at": "2019-08-24T14:15:22Z", "latest": true, "latest_source": "auto", "latest_window_days": 0 }, "commercial": { "user_charge_policy": "paid", "free_reason": "platform", "free_until": "2019-08-24T14:15:22Z" }, "badges": [ { "key": "free", "tone": "green", "priority": 0, "tooltip": { "key": "string", "values": { "property1": "string", "property2": "string" } } } ], "request_format_summary": { "public_operations": [ "string" ], "request_endpoint": "string", "request_endpoint_by_operation": { "property1": "string", "property2": "string" } } } } ]}Authorization
BearerAuth API Key authentication. Create or manage API keys in Dashboard > API > API Keys.
In: header
Query Parameters
Filter by public model provider, for example openai, anthropic, google, or minimax.
Filter by a capability tag such as reasoning, tool-use, vision, or text-to-image. The values in a model's tokenlab.capabilities are valid tags. To filter by kind of model, use category.
Filter by model category.
Value in
- "chat"
- "embedding"
- "translation"
- "image"
- "video"
- "audio"
- "tts"
- "stt"
- "rerank"
- "3d"
- "music"
- "decision"
Sort supported non-chat models for a task category and include recommendation evidence.
Value in
- "image"
- "video"
- "music"
- "3d"
- "tts"
- "stt"
- "embedding"
- "rerank"
- "translation"
- "decision"
Choose compact model-selection fields or the full OpenAI-compatible list shape.
"full"Value in
- "compact"
- "full"
Response
application/json
application/json
application/json
application/json
application/json