Whisper Large v3 Turbo API
Available NewWhisper Large v3 Turbo keeps Large v3's encoder and cuts the decoder from 32 layers to 4, so it transcribes noticeably faster with a small accuracy cost. It runs on our own GPU with batched inference: in our 20 November 2026 test a 61.7-minute MP3 came back in 17.5 seconds end to end, upload included. The model id is the one Groq uses, so Groq code works by changing the base URL and key.
Quickstart
Set INFERENCE_API_KEY to your key from the API Keys page, then run:
curl https://api.inferenceapis.com/v1/audio/transcriptions \
-H "Authorization: Bearer $INFERENCE_API_KEY" \
-F "model=openai/whisper-large-v3-turbo" \
-F "file=@meeting.mp3" \
-F "response_format=verbose_json"
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.inferenceapis.com/v1",
api_key=os.environ["INFERENCE_API_KEY"],
)
with open("meeting.mp3", "rb") as f:
result = client.audio.transcriptions.create(
model="openai/whisper-large-v3-turbo",
file=f,
response_format="verbose_json",
)
print(result.text)
import fs from "node:fs";
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.inferenceapis.com/v1",
apiKey: process.env.INFERENCE_API_KEY,
});
const result = await client.audio.transcriptions.create({
model: "openai/whisper-large-v3-turbo",
file: fs.createReadStream("meeting.mp3"),
response_format: "verbose_json",
});
console.log(result.text);
When to use it
- Long files and batch jobs where speed matters more than the last bit of accuracy
- Code written for Groq's whisper-large-v3-turbo id
- An hour of audio in about 17 seconds for $0.06
- Hard audio or rare languages, where full Whisper Large v3 is more accurate
- You need live streaming transcription
What it costs in practice
| Workload | Cost |
|---|---|
| 1 hour of audio | $0.06 |
| 100 hours of audio | $6.00 |
| 1,000 hours of audio | $60.00 |
Computed from the live rate below. There is no subscription or minimum; new accounts start with free credit.
Endpoint
Content-Type: multipart/form-data
/openai/v1/… is accepted too, so Groq-style base URLs work with only the host changed.
Model IDs and aliases
Send any of these as model; they all resolve to this model and bill at its rate.
| ID | Note |
|---|---|
openai/whisper-large-v3-turbo | Canonical |
whisper-large-v3-turbo | Same id as on Groq — switch by changing the base URL only |
whisper-turbo | Alias |
large-v3-turbo | Alias |
Request body
| Parameter | Type | Description |
|---|---|---|
model required |
string | Model ID: openai/whisper-large-v3-turbo — also accepted: whisper-large-v3-turbo, whisper-turbo, large-v3-turbo |
file required |
file | The audio file (multipart/form-data): mp3, mp4, m4a, wav, webm, flac, ogg. Up to 100 MB. |
language |
string | ISO-639-1 language code, e.g. en. Improves accuracy and latency when known. |
prompt |
string | Optional text to guide style or spelling of names and terms. |
response_format |
string | json (default), verbose_json, text, srt or vtt. |
timestamp_granularities |
array | ["segment"] and/or ["word"] with verbose_json. |
temperature |
number | Sampling temperature between 0 and 1. |
Response
{
"language": "en",
"duration": 12.4,
"text": "Thanks everyone for joining. Let's start with the roadmap.",
"segments": [
{ "id": 0, "start": 0.0, "end": 3.1, "text": "Thanks everyone for joining." },
{ "id": 1, "start": 3.1, "end": 6.0, "text": "Let's start with the roadmap." }
]
}
Pricing
| Price | $0.001 / audio minute |
| Billing | Pay as you go from prepaid credits. No subscription, no daily request or token cap. See all pricing. |
Notes
- Billing is per minute of audio (measured from the uploaded file). Files up to 100 MB are accepted.
- With
diarize=truethe request runs on full Whisper Large v3 on a hosted backend, at $0.002 per minute. - If our GPU is unavailable the request falls back to Whisper Large v3 on a hosted backend at the same price.
Compare Whisper Large v3 Turbo with…
Languages
Per-language codes, tips and quickstarts: speech-to-text languages.
Errors
Errors use the OpenAI envelope: {"error": {"message", "type", "code"}}. 401 missing or invalid key · 402 insufficient_balance · 404 model_not_found · 503 model_unavailable · 502 backend error, safe to retry. Full table with what to retry: API documentation. Errors from other providers: provider error reference.
