What GPU do I need to run primeline/whisper-large-v3-turbo-german?
809M parameters, published in BF16. View on Hugging Face
whisper-large-v3-turbo-german is published by primeline on Hugging Face, with 392,718 downloads and 64 likes to date. It's a WhisperForConditionalGeneration 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 | 1.5 GB | 1.8 GB | RTX 3070 | 1 | $0.088/hr |
| FP8 (quantized) | 0.8 GB | 0.9 GB | RTX 5060 Ti | 1 | $0.180/hr |
| INT4 (quantized) | 0.4 GB | 0.5 GB | RTX 3070 | 1 | $0.088/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-turbo-german at its published (BF16) precision: 1× RTX 3070, 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-turbo-german: common questions
How much VRAM does whisper-large-v3-turbo-german need?
1.8 GB at BF16, 0.9 GB at FP8 (quantized), 0.5 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 1.5 GB of weights plus inference overhead is the whole requirement.
How many copies of whisper-large-v3-turbo-german fit on one RTX 3070?
4, by VRAM alone. That card carries 8.0 GB and one copy needs 1.8 GB at BF16, 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.
Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More Whisper models
- whisper-large-v3-turbo (809M, F16)
- Whisper-Hindi2Hinglish-Apex (809M, BF16)
- whisper-large-v3-turbo-armenian (809M, F32)
- kazakh-whisper-large-v3-turbo (809M, F16)
- whisper-large-v3-turbo (809M, F32)
Lineage
Based on whisper-large-v3-german.
Alternatives at this size
Other models for automatic-speech-recognition within about a third of whisper-large-v3-turbo-german's 809M parameters, from other model lines.
- parakeet-ctc-1.1b (1.1B, F32)
- Qwen3-ASR-0.6B (938M, BF16)
- nemotron-3.5-asr-streaming-0.6b (638M, F32)
- distil-large-v3 (756M, F16)
- mms-1b-all (965M, F32)
More on whisper-large-v3-turbo-german
Fits on an 8 GB GPU at BF16: every model that fits in 8 GB.
Related reading: RTX 3070 pricing and specs, Whisper variants compared, What AI inference is, and What AI model fine-tuning is.
Your own GPU cloud accounts
One console for the GPU clouds you already rent from. Your keys stay on your machine.
$20 a month, flat