Router DeskHomeWhisper Large V3 Turbo
Whisper Large V3 Turbo
This is a transcription model, not a chat model. The slug is openai/whisper-large-v3-turbo.
Updated 2026-09-27
Whisper Large V3 Turbo turns speech into text. OpenRouter’s model page on 27 Sep 2026 showed a per-second price, not a chat token price. DeepInfra was $0.000003 per second. Groq was $0.000011 per second. The page’s short answer for the cost question was $0.000003 per second, matching the cheaper row.
An hour of audio at the DeepInfra rate is 3,600 × $0.000003 = $0.0108. At the Groq rate it is about $0.04. That is the meter if the page’s per-second unit is what you are billed. Confirm the unit on the model page when you start, because speech endpoints have been quoted both per second and against an internal token count.
Completion price is not the thing to optimize. The audio goes in. Text comes out. Do not send this slug to a chat completions tutorial that expects a message array unless the client documents audio input.
Questions
How much is Whisper Large V3 Turbo?
The model page on 27 Sep 2026 listed $0.000003 per second on DeepInfra and $0.000011 per second on Groq.
What is the model id?
openai/whisper-large-v3-turbo
Is this the same as GPT-4o transcribe?
No. It is the Whisper Turbo checkpoint. Other speech models on OpenRouter have their own ids and rates.
Related
- OpenRouter API URL
The OpenRouter base URL is https://openrouter.ai/api/v1. Chat calls go to /chat/completions with a Bearer key.
- GPT-4o pricing
GPT-4o on OpenRouter was $2.50 input and $10 output per million tokens on 27 Sep 2026. The May 2024 snapshot was $5 / $15.
- Image router
OpenRouter’s image collection includes Gemini and GPT image slugs. Video such as Seedance is billed per second, not as a chat completion.
- OpenRouter pricing and credits
OpenRouter credits are a prepaid dollar balance. Buying credits costs 5.5% ($0.80 minimum). Inference uses each provider’s own token price.