What GPU do I need to run openai/whisper-large-v3?
A 1.5B-parameter speech model. 1.5B parameters, published in F16. View on Hugging Face
whisper-large-v3 is published by openai on Hugging Face, with 5,030,454 downloads and 6,209 likes to date. It's a WhisperForConditionalGeneration model built for automatic-speech-recognition, published natively in F16.
What whisper-large-v3 is
whisper-large-v3 is a 1.5B-parameter model published by OpenAI on Hugging Face that transcribes speech to text, released under Apache 2.0.
License note: fully permissive, no gating and no usage restrictions. Facts in this section are sourced from whisper-large-v3's Hugging Face model card, not benchmarked by Aquanode.
What it's used for
- Speech transcription
- Voice agent pipelines
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) |
|---|---|---|---|---|---|
| FP16 | 2.9 GB | 3.4 GB | V100 | 1 | $0.088/hr |
| FP8 (quantized) | 1.4 GB | 1.7 GB | RTX 5060 Ti | 1 | $0.110/hr |
| INT4 (quantized) | 0.7 GB | 0.9 GB | RTX 5060 Ti | 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 whisper-large-v3 at its published (F16) precision: 1× V100, at $0.088/hr per GPU ($0.088/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
whisper-large-v3: common questions
How much VRAM does whisper-large-v3 need?
3.4 GB at FP16, 1.7 GB at FP8 (quantized), 0.9 GB at INT4 (quantized). All of those sit under 6 GB, the smallest capacity this page reasons about, so VRAM is not what limits where this model runs. The 2.9 GB of weights plus inference overhead is the whole requirement.
How many copies of whisper-large-v3 fit on one V100?
4, by VRAM alone. That card carries 16.0 GB and one copy needs 3.4 GB at FP16, on a live rate of $0.088/hr for the whole card. Throughput is not modelled here, so 4 copies is not 4 times the requests served.
How to run whisper-large-v3
Run whisper-large-v3 with Transformers (Python)
Generic example using Hugging Face's transformers library, not from the model's own docs.
from transformers import pipeline
asr = pipeline("automatic-speech-recognition", model="openai/whisper-large-v3", device="cuda")
result = asr("audio.wav")
print(result["text"])Deploy whisper-large-v3 on Aquanode
Aquanode has no one-click deploy template for whisper-large-v3; you install the inference engine yourself with the commands below. Aquanode sells GPU pods billed per second, not a hosted inference API.
- Launch a bare GPU pod sized to the requirement above (1× V100 or larger).
- Open a terminal on the pod, or save one of the commands above as a startup script so it runs automatically the first time the pod boots.
- Run the command and connect to the resulting endpoint.
Launch a GPU podWeight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More Whisper models
- Whisper-Hindi2Hinglish-Prime (1.5B, F32)
- whisper-large-v3 (1.5B, F32)
- whisper-large-v3-german (1.5B, BF16)
- whisper-large-v3-russian (1.5B, BF16)
- nb-whisper-large (1.5B, F32)
Related reading: V100 pricing and specs, and The best GPUs for AI, ranked.
