API reference · Models
List catalog models (public ones; with credentials, also the project's private ones)
/v1/catalog/models- TypeScript
client.models.list()- Python
client.models.list()- CLI
hayonix models list
Works without credentials. Each model carries its prices from the price book in force, its context length, capabilities and whether a replica is serving it right now. With an API key, or a session plus X-Project-Id, the list also includes that project's private models (fine-tuned models and dedicated endpoints).
None: this operation is public.
Examples
The samples read API_KEY and API_BASE_URL from the environment (see Setup). Values are examples.
curl -sS "$API_BASE_URL/catalog/models" \
-H "Authorization: Bearer $API_KEY"Request
Query parameters
limitinteger1–200, default 50
1–200
cursorstring`next_cursor` of the previous page
up to 200 characters
typestringOnly models of this type
One of
chatcompletionembeddingrerankimagespeech_to_texttext_to_speech
Response
200 a page of results — A page of catalog models
The items of this page.
ShowHide 15 fields of
dataobjectstringrequiredAlways `catalog_model`: the type of this object
One of
catalog_modelidstringrequiredUnique ID (`mdl_…`)
ID with prefix mdl_
modelstringrequiredModel ID (`publisher/name`)
display_namestringrequiredName to show people
typestringrequiredWhat the model does: `chat` (chat completions), `completion` (text completions), `embedding` (vectors), `rerank` (document reranking), `image` (image generation), `speech_to_text` (transcription), `text_to_speech` (speech synthesis), `adapter` (a fine-tuned adapter served on its base model)
One of
chatcompletionembeddingrerankimagespeech_to_texttext_to_speechadaptervisibilitystringrequired`public`: listed for everyone; `private`: visible only to the project that owns it (fine-tuned models and dedicated endpoints)
One of
publicprivatepublisherstring | nullrequiredWho published the model; null when not recorded
descriptionstring | nullrequiredFree-text description
licensestring | nullrequiredThe model's license; null when not recorded
context_lengthinteger | nullrequiredMaximum number of tokens in one request, prompt and output together; null when it does not apply or is not known
capabilitiesobjectrequiredFeatures the model supports, keyed by name (for example `streaming: true`); a missing key means the feature is not offered
statusstringrequired`available`: offered normally; `deprecated`: still listed but no longer recommended and may be removed, so plan a move to another model
One of
availabledeprecatedpricingModelPricing | any JSON | nullrequiredPrices from the price book in force; null when the model has no price
ShowHide 8 fields of
pricingcurrencystringrequiredAlways `usd`
One of
USDunitstringrequiredUnit of the token prices: nano-USD (1 USD = 1_000_000_000) per 1 million tokens
One of
nano_usd_per_1m_tokensinputstring | nullrequirednano-USD per 1M tokens, as a decimal string
pattern ^-?[0-9]+$
cached_inputstring | nullnano-USD per 1M tokens, as a decimal string
pattern ^-?[0-9]+$
outputstring | nullnano-USD per 1M tokens, as a decimal string
pattern ^-?[0-9]+$
placeholderbooleanrequiredTrue while the prices are development placeholders
batchobject | nullPrices for the batch API (asynchronous, discounted), per 1M tokens; null when the model cannot be used in a batch
ShowHide 3 fields of
batchinputstring | nullrequirednano-USD per 1M tokens, as a decimal string
pattern ^-?[0-9]+$
cached_inputstring | nullnano-USD per 1M tokens, as a decimal string
pattern ^-?[0-9]+$
outputstring | nullnano-USD per 1M tokens, as a decimal string
pattern ^-?[0-9]+$
unitsarray of objectSpeech-to-text, text-to-speech and image models are priced per unit, not per token: one entry per meter (the token prices are then null)
ShowHide 4 fields of
unitsmeterstringrequiredWhat is counted: `audio_input_ms` (milliseconds of input audio), `speech_character` (input characters), `image_small` / `image_medium` / `image_large` (generated images by size tier)
unit_pricestringrequirednano-USD per `unit_quantity` units
pattern ^-?[0-9]+$
unit_quantitystringrequiredDecimal string
display_unitstring | nullrequiredHuman-readable unit that `unit_price` is quoted in; null when there is none
servingbooleanrequiredWhether a replica is ready to serve the model right now (the same sources the inference gateway routes by); a listed model may have none
created_atstring (date-time)requiredWhen it was created (ISO 8601, UTC)
next_cursorstring | nullrequiredPass as `cursor` for the next page; null on the last page.
Example response
{
"data": [
{
"object": "catalog_model",
"id": "mdl_01jc5zq4m8k3v6x9t2w7r0b1e4",
"model": "qwen/qwen2.5-0.5b-instruct",
"display_name": "string",
"type": "chat",
"visibility": "public",
"publisher": "string",
"description": "string",
"license": "string",
"context_length": 1,
"capabilities": {},
"status": "available",
"pricing": {
"currency": "USD",
"unit": "nano_usd_per_1m_tokens",
"input": "100000000",
"cached_input": "100000000",
"output": "100000000",
"placeholder": true,
"batch": {
"input": "100000000",
"cached_input": "100000000",
"output": "100000000"
},
"units": [
{
"meter": null,
"unit_price": null,
"unit_quantity": null,
"display_unit": null
}
]
},
"serving": true,
"created_at": "2026-01-01T00:00:00.000Z"
}
],
"next_cursor": "string"
}Errors
Errors use the standard envelope. What this operation can answer:
| Status | Type | When |
|---|---|---|
400 | invalid_request_error | A parameter or body field is missing or not allowed; param names it. |
429 | rate_limit_error rate_limited | Too many requests; retry after retry-after seconds. |