GET

/v1/models/replicate/replicate-ibm-granite-granite-speech-4-1-2b

Accept:

provider replicatestatus activefirst seen Apr 27, 2026last seen 9m ago

rawId	ibm-granite/granite-speech-4.1-2b
displayName	granite-speech-4.1-2b
id	replicate-ibm-granite-granite-speech-4-1-2b
activity	chat
schemaEndpointId	models/ibm-granite/granite-speech-4.1-2b/predictions input · output
releasedAt	1777332447 — Apr 27, 2026
firstSeenAt	1777332447 — Apr 27, 2026
lastSeenAt	1794561526 — 9m ago
pricing	$0.03 / 1M input · $0.25 / 1M output → open calculator https://replicate.com/ibm-granite/granite-speech-4.1-2b input_tokens (usage) · output_tokens (usage)
modalities	{"input":["text","audio"],"output":["text"]}
capabilities	{"frequency_penalty":true,"max_tokens":true,"presence_penalty":true,"seed":true,"stop":true,"temperature":true,"top_k":true,"top_p":true}
providerMetadata	{"visibility":"public","official":true}
contextWindow · maxOutput · knowledgeCutoff · openWeights · weightsUrl · requestMap	not published by this provider’s list endpoint

Endpoints & schemas

GET /v1/schemas/replicate →

endpoint	activity
models/ibm-granite/granite-speech-4.1-2b/predictions input · output	chat

endpoints matching this model — the provider has more, see /v1/schemas/replicate

Sources

field	recorded source
modalities	https://api.replicate.com/v1/models/ibm-granite/granite-speech-4.1-2b listing /latest_version/openapi_schema/components/schemas/Input/properties
pricing	Source linked above. docs-derived billingConfig Source trace `9ed5e97c5c58523ee3cc018c7bb818b1dd8cd3da349df74c26571c5c35df65a6`
capabilities.frequency_penalty	https://api.replicate.com/v1/models/ibm-granite/granite-speech-4.1-2b listing /latest_version/openapi_schema/components/schemas/Input/properties/frequency_penalty
capabilities.max_tokens	https://api.replicate.com/v1/models/ibm-granite/granite-speech-4.1-2b listing /latest_version/openapi_schema/components/schemas/Input/properties/max_tokens
capabilities.presence_penalty	https://api.replicate.com/v1/models/ibm-granite/granite-speech-4.1-2b listing /latest_version/openapi_schema/components/schemas/Input/properties/presence_penalty
capabilities.seed	https://api.replicate.com/v1/models/ibm-granite/granite-speech-4.1-2b listing /latest_version/openapi_schema/components/schemas/Input/properties/seed
capabilities.stop	https://api.replicate.com/v1/models/ibm-granite/granite-speech-4.1-2b listing /latest_version/openapi_schema/components/schemas/Input/properties/stop
capabilities.temperature	https://api.replicate.com/v1/models/ibm-granite/granite-speech-4.1-2b listing /latest_version/openapi_schema/components/schemas/Input/properties/temperature
capabilities.top_k	https://api.replicate.com/v1/models/ibm-granite/granite-speech-4.1-2b listing /latest_version/openapi_schema/components/schemas/Input/properties/top_k
capabilities.top_p	https://api.replicate.com/v1/models/ibm-granite/granite-speech-4.1-2b listing /latest_version/openapi_schema/components/schemas/Input/properties/top_p
cacheRead	https://replicate.com/pricing Value unknown (null); not published in the checked source · checked 2026-10-06.
requestMap	https://api.replicate.com/openapi.json Value unknown (null); not published in the checked source · checked 2026-10-06.
activity	Source unknown — no source URL recorded.
contextWindow	Source unknown — no source URL recorded.
maxOutput	Source unknown — no source URL recorded.
reasoning	Source unknown — no source URL recorded.
schemaEndpointId	Source unknown — no source URL recorded.

Use it

shell

$ curl https://modelschemas.com/v1/schemas/replicate/chat/models/ibm-granite/granite-speech-4.1-2b/predictions # request schema
$ curl -X POST https://modelschemas.com/v1/validate -d '{"provider":"replicate","endpointId":"models/ibm-granite/granite-speech-4.1-2b/predictions","payload":{…}}'
$ curl -X POST https://modelschemas.com/v1/estimate -d '{"provider":"replicate","model":"ibm-granite/granite-speech-4.1-2b","usage":{"input_tokens":1200,"output_tokens":400}}'