Replicate
Replicate
Multimodal

Whisper Large v3

Audio

OpenAI Whisper Large v3 hosted by Replicate. Default open-weights speech-to-text model for production transcription.

Whisper Large v3 is a multimodal AI model from Replicate. It costs $0.004 per million input tokens.

INPUT
$0.004/M
per million input tokens
OUTPUT
—
per million output tokens
BLENDED 70/30
—
unchanged since 3 May
CONTEXT
—
no context window

Benchmarks

No published benchmark scores tracked for this model yet. Frontier and reasoning models from major providers have scores; smaller models and inference-host variants typically inherit the underlying open-weights score.

More from Replicate

See all 5 →

Frequently asked questions

How much does Whisper Large v3 cost?

Whisper Large v3 costs $0.004 per million input tokens.

What is Whisper Large v3 best for?

OpenAI Whisper Large v3 hosted by Replicate. Default open-weights speech-to-text model for production transcription.

Who makes Whisper Large v3?

Whisper Large v3 is developed and served by Replicate.

Terms used on this page