Model reference · open weights
whisper-large-finnish is an open-weight audio or speech model from Finnish-NLP. whisper-large-finnish-v3 (FP16) weighs 3.2 GB; the smallest configuration that runs it is RTX 3060 12 GB.
For self-hosting, compare this build's requirements below with the GPU you plan to rent.
Summary of the Finnish-NLP/whisper-large-finnish-v3 model card, 2026-10-01
What it is
| Released by | Finnish-NLP |
|---|---|
| Released | 2023-11-25 |
| Parameters | 1.6B |
| VRAM | 3.2 GB for the weights |
What it runs on
| Card | One stream | Memory |
|---|---|---|
| RTX 3060 12 GB | fits | 11.6 GB |
| RTX 4060 Ti 16 GB | fits | 15.4 GB |
| RTX 3090 24 GB | fits | 23.4 GB |
| RTX 4090 24 GB | fits | 23.4 GB |
| RTX 5090 32 GB | fits | 31.0 GB |
| L40S 48 GB | fits | 44.0 GB |
| A100 80 GB | fits | 78.2 GB |
| H100 80 GB | fits | 78.1 GB |
| RTX PRO 6000 Blackwell 96 GB | fits | 93.8 GB |
| DGX Spark (GB10) 128 GB unified | fits | 107 GB |
| H200 141 GB | fits | 138 GB |
| B200 180 GB | fits | 176 GB |
From the model card
CV11
Fleurs
WER: 12.04
WER NORMALIZED: 9.63
CER: 2.48
CER NORMALIZED: 3.64
@14000 finetuning steps
CV11
Fleurs
@32000 finetuning steps
CV11
Fleurs
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.
Benchmarks
As published on the model card: the maker's own numbers, not measured by AxForge.
| Task | Dataset | Metric | Score |
|---|---|---|---|
| Automatic Speech Recognition | Common Voice 11.0 | Wer | 8.230 |
| Automatic Speech Recognition | Common Voice 11.0 | Cer | 1.430 |
| Automatic Speech Recognition | FLEURS | Wer | 8.210 |
| Automatic Speech Recognition | FLEURS | Cer | 3.230 |