best model by language
Benchmarked on the FLEURS test split at Q8_0. Error is WER — lower is better. Speed is × realtime on Handy target hardware.
accuracy vs speed
Malay · 11 models
AMD Ryzen 4750U GPU (Vulkan)
pareto frontier (no model is both faster and more accurate) other models
highlighted rows are recommended picks or on pareto frontier
| | | | | | |
| 1 | whisper-large-v3 | 7.59 | 2.0× | 1.0× | 1.55 GB | Apache-2.0 |
| 2 | whisper-large-v3-turboaccurate pick | 8.64 | 3.1× | 1.4× | 0.83 GB | Apache-2.0 |
| 3 | whisper-large-v2 | 9.38 | 2.0× | 1.0× | 1.55 GB | Apache-2.0 |
| 4 | Fun-ASR-MLT-Nano-2512!overall pickfast pick | 9.92 | 8.8× | 5.5× | 0.83 GB | FunASR Model Open Source License Agreement v1.1 |
| 5 | Qwen3-ASR-1.7B | 10.42 | 4.1× | 2.5× | 2.04 GB | Apache-2.0 |
| 6 | whisper-large | 11.28 | 2.0× | 1.0× | 1.55 GB | Apache-2.0 |
| 7 | whisper-medium | 13.23 | 4.1× | 1.9× | 0.77 GB | Apache-2.0 |
| 8 | Qwen3-ASR-0.6B | 17.18 | 9.7× | 5.3× | 0.79 GB | Apache-2.0 |
| 9 | whisper-small | 21.48 | 11.7× | 5.9× | 0.25 GB | Apache-2.0 |
| 10 | whisper-base | 40.87 | 29.0× | 19.3× | 0.08 GB | Apache-2.0 |
| 11 | whisper-tiny | 62.17 | 43.1× | 39.8× | 0.04 GB | Apache-2.0 |