What GPU do I need to run microsoft/VibeVoice-ASR?
8.7B parameters, published in BF16. View on Hugging Face
Set up VibeVoice-ASRVibeVoice-ASR is published by microsoft on Hugging Face, with 699,615 downloads and 1,279 likes to date. It's a VibeVoiceForASRTraining model built for automatic-speech-recognition, published natively in BF16.
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision, plus a fixed overhead for activations and allocator fragmentation. Speech models don't build the same growing KV-cache a text model does. Memory scales primarily with input audio length. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 16.2 GB | 19.4 GB | RTX A5000 | 1 | $0.176/hr |
| FP8 (quantized) | 8.1 GB | 9.7 GB | RTX 4070 Super | 1 | $0.121/hr |
| INT4 (quantized) | 4.0 GB | 4.8 GB | RTX 3060 | 1 | $0.110/hr |
A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.
INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.
Cheapest way to run VibeVoice-ASR at its published (BF16) precision: 1× RTX A5000, at $0.176/hr per GPU ($0.176/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
VibeVoice-ASR: common questions
Does VibeVoice-ASR fit on a 24 GB GPU?
Yes. At BF16 it needs 19.4 GB of VRAM, so a 24 GB card holds it with 4.6 GB to spare. A 16 GB card is not enough for it at BF16.
What is the least VRAM VibeVoice-ASR can run in?
4.8 GB, at INT4 (quantized), which fits a 6 GB card, against 19.4 GB at BF16. That row assumes an INT4 checkpoint has actually been published for this model, so check its Hugging Face page before planning around it.
Does quantizing VibeVoice-ASR lower the GPU bill?
Yes. At BF16 the cheapest live fit is one RTX A5000 at $0.176/hr. At INT4 (quantized) it drops to one RTX 3060 at $0.110/hr, provided a quantized checkpoint exists for it.
Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More VibeVoice models
- VibeVoice-ASR-BitNet (323M, F32)
- VibeVoice-ASR-HF (8.3B, BF16)
- VibeVoice-1.5B (2.7B, BF16)
Alternatives at this size
Other models for automatic-speech-recognition within about a third of VibeVoice-ASR's 8.7B parameters, from other model lines.
- granite-speech-3.2-8b (8.5B, BF16)
- higgs-audio-v3-8b-stt-v2 (8.9B, BF16)
More on VibeVoice-ASR
Fits on a 24 GB GPU at BF16: every model that fits in 24 GB.
Fits on a 12 GB GPU at FP8: every model that fits in 12 GB.
Fits on an 8 GB GPU at INT4: every model that fits in 8 GB.
Best speech-to-text models: how VibeVoice-ASR ranks against the rest.
Related reading: RTX A5000 pricing and specs, Whisper variants compared, and What AI inference is.
