What GPU do I need to run ServiceNow-AI/Apriel-Voice-27B-preview?
27.2B parameters, published in BF16. View on Hugging Face
Set up Apriel-Voice-27B-previewApriel-Voice-27B-preview is published by ServiceNow-AI on Hugging Face, with 915 downloads and 90 likes to date. It's a AprielVoiceModel model built for audio-text-to-text, published natively in BF16.
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision times a flat 1.2 overhead for activations and fragmentation. The KV-cache grows with context and is not in that factor; it is listed per model below where the architecture is published. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 50.7 GB | 60.9 GB | A100 | 1 | $1.18/hr |
| cheaper alt. | RTX 3090 | 3 | $0.485/hr | ||
| FP8 (quantized) | 25.4 GB | 30.4 GB | RTX 4080 Super | 1 | $0.361/hr |
| cheaper alt. | RTX 5060 Ti | 2 | $0.330/hr | ||
| INT4 (quantized) | 12.7 GB | 15.2 GB | RTX 3090 | 1 | $0.162/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 Apriel-Voice-27B-preview at its published (BF16) precision: 1× A100, at $1.18/hr per GPU ($1.18/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
Apriel-Voice-27B-preview: KV cache by context length
The KV cache is the memory the attention layers hold for every token of context, on top of the weights. It is not part of the flat 1.2x overhead in the table above, grows with context length and with every concurrent request, and is why a long-context deployment needs more VRAM than the table shows.
KV-cache size for Apriel-Voice-27B-preview is not published for this architecture, so no figure is shown.
Apriel-Voice-27B-preview: common questions
Can Apriel-Voice-27B-preview run on a single GPU?
Yes, but not on a desktop card. At BF16 it needs 60.9 GB of VRAM, more than a single 32 GB desktop card holds. The cheapest live single-card fit is a 80.0 GB A100 at $1.18/hr.
What is the least VRAM Apriel-Voice-27B-preview can run in?
15.2 GB, at INT4 (quantized), which fits a 16 GB card, against 60.9 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 Apriel-Voice-27B-preview lower the GPU bill?
Yes. At BF16 the cheapest live fit is one A100 at $1.18/hr. At INT4 (quantized) it drops to one RTX 3090 at $0.162/hr, provided a quantized checkpoint exists for it.
Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More on Apriel-Voice-27B-preview
Fits on an 80 GB GPU at BF16: every model that fits in 80 GB.
Fits on a 32 GB GPU at FP8: every model that fits in 32 GB.
Fits on a 16 GB GPU at INT4: every model that fits in 16 GB.
Related reading: A100 pricing and specs, How much VRAM you need for LLMs, Serving LLMs with vLLM, vLLM vs TensorRT-LLM vs SGLang, and Best GPU for LLM inference.
