Whisper Large v3 Turbo
AvailableWhisper Large v3 Turbo is the speed-optimized speech-to-text model of the OpenAI Whisper family. It transcribes 99+ languages and is used through the OpenAI Audio API-compatible /v1/audio/transcriptions endpoint.
Quickstart
OpenAI-compatible API. Point base_url at the gateway and pass the Model ID in the model field.
curl https://llm.fixcloud.ai/v1/audio/transcriptions \
-H "Authorization: Bearer $FXC_API_KEY" \
-F file="@meeting.mp3" \
-F model="whisper-large-v3-turbo" \
-F language="tr"
from openai import OpenAI
client = OpenAI(base_url="https://llm.fixcloud.ai/v1", api_key="FXC_API_KEY")
with open("meeting.mp3", "rb") as audio:
transcript = client.audio.transcriptions.create(
model="whisper-large-v3-turbo",
file=audio,
)
print(transcript.text)
import fs from 'node:fs';
import OpenAI from 'openai';
const client = new OpenAI({ baseURL: 'https://llm.fixcloud.ai/v1', apiKey: process.env.FXC_API_KEY });
const transcript = await client.audio.transcriptions.create({
model: 'whisper-large-v3-turbo',
file: fs.createReadStream('meeting.mp3'),
});
console.log(transcript.text);
Capabilities
The features this model offers through the API.
- Audio inputAccepts audio input.
Pricing
Prices are per 1 million tokens. Once the monthly token quota of your plan is used up, usage is billed at these prices.
- Per request
- —
- Per day
- —
- Per month (30 days)
- —
Enter your expected usage; actual billing follows the token counts measured by the gateway.
View plansSpecifications
Transparency details about the infrastructure serving this model.
- Family
- whisper
- Publisher
- OpenAI
- License
- mit
- Model ID
- whisper-large-v3-turbo
- Quantization
- None (native)
- GPUs per replica
- 1 × GPU
Up to 25 MB of audio per file (about 25 minutes of mp3).