4 GB accelerator memory
FP16 · whisper.cpp / Transformers
GPU acceleration is the practical provider path for throughput and long audio queues.
OpenAI
Audio model
Whisper Large V3 Turbo
A multilingual speech recognizer optimized for faster transcription.
Transcription
Capabilities
What you can use it for.
The most useful reasons to choose this model, without making you read through its repository first.
TranscriptionTranslationMultilingual audioLong-form speech
01
Meeting transcripts
02
Subtitles
03
Multilingual audio
For providers
Want to serve this model?
The provider requirements live here, separate from the information you need to choose and use the model.
Starting pointCPU · 8 GB RAM
Conservative estimate
CPU + 8 GB RAM
FP32 / quantized · whisper.cpp
The official card supports CPU fallback; real-time throughput is not guaranteed.
Source and license
Check the original model.
OpenMayhem links back to the source repository so you can inspect the model card, files, license, limitations, and creator guidance directly.
