Skip to main content
GET
List audio models
Discovery endpoint for speech, transcription, music and sound effects. Filter by kind with ?type=tts, ?type=stt, ?type=music or ?type=sound.

Query Parameters

type
enum<string>

Narrow the audio family to tts, stt, music or sound. Default: all of them.

Available options:
tts,
stt,
music,
sound
include
string

Comma-separated list of optional fields to attach to each item. Supported: maker (public), route (requires API key), benchmarks (public), policy (requires API key — per-item comply allowlist verdict), editorial (public: the model's short_description and long_description).

Example:

"maker"

q
string

Case-insensitive substring of a model's id, name, description or alias. Dots, dashes, underscores, colons, slashes and spaces are also ignored on both sides from the start of a word, so gpt-5-1, gpt 5.1 and gpt5.1 all find gpt-5.1.

Example:

"claude-sonnet"

provider
string

Filter by provider name (e.g. openai, xai).

deprecated
boolean

Set to true to include retired models in results.

sort
string
default:id

Field to sort by (id, name, provider, cost, speed, quality). Default id.

order
string
default:asc

asc (default) or desc.

offset
integer
default:0

Pagination offset.

Required range: x >= 0
limit
integer
default:50

Page size. A non-positive value returns all rows from offset.

Response

Successful response

has_more
boolean
required

True when more rows follow this page. Request the next page with offset set to next_offset.

limit
integer
required
models
object[]
required
offset
integer
required
total
integer
required
next_offset
integer

Offset of the next page. Present only when has_more is true.

Last modified on October 1, 2026