OpenAIWhisper Large V3 on NVIDIA Grace Blackwell Ultra GB300
Yes — Grace Blackwell Ultra GB300 runs Whisper Large V3 excellently at FP16 — 450 tok/s. 288 GB VRAM with plenty of headroom.
Model Size
1.55B
Device VRAM
288 GB
Bandwidth
8000 GB/s
Quantization
FP16
Performance by Quantization
OwnRig currently has one published compatibility entry for Whisper Large V3 on NVIDIA Grace Blackwell Ultra GB300 at FP16. This is the best supported pairing we can stand behind today.
| Quantization | Speed | TTFT | Fits in VRAM | Rating | Confidence |
|---|---|---|---|---|---|
| FP16 | 450 tok/s | 15ms | ✓ Yes | Excellent | estimated |
Notes
FP16
Full FP16 quality transcription. 285GB headroom. Real-time+ processing.
About Whisper Large V3
Whisper Large V3 (1.55B) is a transcription model. OpenAI's best open speech-to-text model. Supports 99 languages. Near-human accuracy for English. Low VRAM requirements; runs on any GPU. Useful for builders who need voice-to-code or meeting transcription.
View all Whisper Large V3 hardware options →About NVIDIA Grace Blackwell Ultra GB300
NVIDIA Grace Blackwell Ultra GB300 has 288 GB at 8000 GB/s. Street price: $30,000.
See all models NVIDIA Grace Blackwell Ultra GB300 can run →Estimate method: Estimated from GB300 bandwidth ratio vs RTX 4090. Reference hardware source: dell.com (2026-03-23)
Performance varies by driver version, inference engine, quantization method, context length, and system configuration. Figures shown are estimates based on community benchmarks and may not reflect your exact setup. Product names are trademarks of their respective owners. OwnRig is independent and not affiliated with any hardware or AI model provider.