Skip to content
Documentation
GuidesAPI ReferenceIntegrations
ArenaPlatform
API Reference/Models and Providers

List Models

GET/v1/models

Overview

Public, as on OpenRouter: no key or scope is needed.

Routing and provider details

A bearer key that is sent is still checked, so a wrong key returns 401 and the list doubles as a connection check. Each client address may make 120 requests a minute to the model list, model lookups, endpoint lists and model stats together, with or without a key; beyond that they return 429.

Language models come first. Each carries OpenRouter's fields alongside the catalog's: created, description, context_length, pricing (USD per token, what AudioLab charges: the provider's rate minus 15%), architecture, supported_parameters (what the model's contract honours; others are dropped or refused as the Chat Completions reference describes), top_provider.max_completion_tokens (the output cap) and expiration_date. Input modalities say what a request may carry, so language models list text only. costream/auto and costream/fusion are listed when routed execution is enabled; their price is "-1" because it is that of the models that answer.

A model that several hosts serve is listed once, under its host-free ID, with the price of its default host (its maker's own when that is a host, otherwise the cheapest), the largest context and output cap any host accepts, every parameter some host supports, and each host in endpoints. Each host's own ID is not listed but still resolves and selects that host alone.

A model that has been switched off, or whose provider shutdown has passed, is left out. A scheduled shutdown shows as expiration_date (YYYY-MM-DD, UTC).

Authorization

Authorizationstring · headerRequired

AudioLab API key sent as Bearer <token>. Authentication guide

Parameters

output_modalitiesquery · string

Comma-separated output modalities. Matches any selected modality.

taskquery · string

Responses

200Model page.
application/json
dataobject[]Required
Show nested fields
idstringRequired
namestringRequired
ownerstringRequired
taskstringRequired

Values: "llm""transcription""speech""s2s""embeddings""moderation""images""videos""ocr""rerank""music""audio""classification""vision"

input_modalitiesstring[]Required
Show nested fields

See the OpenAPI schema for this nested object.

output_modalitiesstring[]Required
Show nested fields

See the OpenAPI schema for this nested object.

statestringRequired

Values: "active""preview""disabled"

languagesstring[]
Show nested fields

See the OpenAPI schema for this nested object.

featuresobject
Show nested fields
batchboolean
streamboolean
diarizationboolean
word_timestampsboolean
segment_timestampsboolean
confidenceboolean
redactionboolean
translationboolean
keytermsboolean
subtitlesboolean
duplexboolean
realtimeboolean
audio_inputboolean
audio_outputboolean
toolsboolean
default_routestring | null
endpointsstring[]Required
Show nested fields

See the OpenAPI schema for this nested object.

expiration_datestring | null

The provider's announced shutdown (UTC date), or null.

disabled_reasonstring

Why a model was switched off; only on a single-model lookup of a withdrawn model.

canonical_slugstring

Language models. The same as id.

createdinteger | null

Language models. Release date as Unix seconds; null when the maker never published one.

descriptionstring
context_lengthinteger

Language models. The most input tokens a request may carry.

architectureobject

Language models.

Show nested fields
modalitystring
input_modalitiesstring[]

What a request may carry. image and file (PDF) parts bill as input tokens.

Show nested fields

See the OpenAPI schema for this nested object.

output_modalitiesstring[]
Show nested fields

See the OpenAPI schema for this nested object.

tokenizerstring
instruct_typenull
pricingobject

Language models. USD per token as decimal strings, what AudioLab charges (the provider's rate minus 15%). "-1" when the price is that of the models that answer.

Show nested fields
promptstring
completionstring
input_cache_readstring
web_searchstring

USD per web search call, query or grounded request with the model's own search; "0" when it cannot search.

top_providerobject

Language models.

Show nested fields
context_lengthinteger
max_completion_tokensinteger | null

The output cap. A request without max_tokens may generate up to it.

is_moderatedboolean
per_request_limitsnull
supported_parametersstring[]

Language models. Request parameters the model's contract honours.

Show nested fields

See the OpenAPI schema for this nested object.

default_parametersobject
supported_endpointsstring[]

Set when a model is served on one API only.

Show nested fields

See the OpenAPI schema for this nested object.

400Request validation failed.
application/json
errorobjectRequired
Show nested fields
codestringRequired
messagestringRequired
paramstring | null
request_idstring | null
metadataobject

Typed detail for some refusals. budget_exceeded (402): limit (key_budget or workspace_budget) and budget with scope, period, limit, usage and resets_at. key_limit_exceeded (402): limit key_spend_limit. access_denied (403): policy (workspace or key), rule (models, providers or products), product, provider and model.

401Authentication is missing or invalid.
application/json
errorobjectRequired
Show nested fields
codestringRequired
messagestringRequired
paramstring | null
request_idstring | null
metadataobject

Typed detail for some refusals. budget_exceeded (402): limit (key_budget or workspace_budget) and budget with scope, period, limit, usage and resets_at. key_limit_exceeded (402): limit key_spend_limit. access_denied (403): policy (workspace or key), rule (models, providers or products), product, provider and model.

429Rate limit exceeded.
application/json
errorobjectRequired
Show nested fields
codestringRequired
messagestringRequired
paramstring | null
request_idstring | null
metadataobject

Typed detail for some refusals. budget_exceeded (402): limit (key_budget or workspace_budget) and budget with scope, period, limit, usage and resets_at. key_limit_exceeded (402): limit key_spend_limit. access_denied (403): policy (workspace or key), rule (models, providers or products), product, provider and model.

500Internal server error.
application/json
errorobjectRequired
Show nested fields
codestringRequired
messagestringRequired
paramstring | null
request_idstring | null
metadataobject

Typed detail for some refusals. budget_exceeded (402): limit (key_budget or workspace_budget) and budget with scope, period, limit, usage and resets_at. key_limit_exceeded (402): limit key_spend_limit. access_denied (403): policy (workspace or key), rule (models, providers or products), product, provider and model.