# Whisper Large v3 API on Opennan

> Whisper Large v3 API on Opennan: $0.000585 per minute (billed per started minute), 90% below OpenAI's official $0.006 per minute (billed per started minute).

- Maker: OpenAI
- Category: Voice
- Model id: `whisper-large-v3`
- Endpoint: POST https://api.opennan.com/v1/audio/transcriptions
- MCP: https://api.opennan.com/v1/mcp (transcribe_audio)
- Page: https://opennan.com/models/whisper-large-v3

## Whisper Large v3 API prices

| Tier | Opennan | Official | Below official |
| --- | --- | --- | --- |
| Per minute of audio, billed per started minute | $0.000585 | $0.006 | 90% below |

Official is the model maker's own list price for the same tier. Where the maker publishes a price, we never sell above it, most tiers at least 10% below.

## Call Whisper Large v3 from code

```bash
curl https://api.opennan.com/v1/audio/transcriptions \
  -H "Authorization: Bearer $OPENNAN_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"whisper-large-v3","audio_url":"https://cdn.opennan.com/examples/audio/3f563c93-04da-490f-b9ff-023ea49e3893_example.mp3","language":"en"}'

# Then poll until "status" is "succeeded" or "failed":
curl https://api.opennan.com/v1/tasks/<task id> -H "Authorization: Bearer $OPENNAN_KEY"
```

## Whisper Large v3 API questions

### How much does the Whisper Large v3 API cost?

Whisper Large v3 API on Opennan: $0.000585 per minute (billed per started minute), 90% below OpenAI's official $0.006 per minute (billed per started minute). Prices are in US dollars per call, shown before you run, and every tier is in the price table.

### How does that compare with the official price?

OpenAI lists $0.006 per minute (billed per started minute); Opennan charges $0.000585 per minute (billed per started minute), 90% below. We only sell a tier below the maker's official price, most tiers at least 10% below.

### What inputs does Whisper Large v3 take?

Audio.

### Which options can a call set?

Songs up to 30 minutes long.

### How is it billed?

Per minute of audio transcribed, billed per started minute: a 61-second file is billed as 2 minutes.

### What happens if a generation fails?

A failed task is refunded in full. You pay only for results.

### How do I call Whisper Large v3 from an AI agent?

Add the MCP server https://api.opennan.com/v1/mcp with your API key as a bearer token, then call the transcribe_audio tool with model "whisper-large-v3". Or send a POST to https://api.opennan.com/v1/audio/transcriptions.

## Similar models

- [Qwen3-ASR 1.7B](https://opennan.com/models/qwen3-asr-1.7b): from $0.000585 per minute (billed per started minute)
- [Nemotron 3.5 ASR](https://opennan.com/models/nemotron-asr): from $0.00026 per minute (billed per started minute)
- [Qwen3-ASR 0.6B](https://opennan.com/models/qwen3-asr-0.6b): from $0.00026 per minute (billed per started minute)
- [Whisper Large v3 Turbo](https://opennan.com/models/whisper-large-v3-turbo): from $0.00026 per minute (billed per started minute)
- [Voxtral Small 24B](https://opennan.com/models/voxtral-small): from $0.0039 per minute (billed per started minute)
- [Kokoro](https://opennan.com/models/kokoro): from $1.287 per 1M characters