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

Persian · 6 models · left is slower, down is better · log speed axis

AMD Ryzen 4750U GPU (Vulkan)
pareto frontier (no model is both faster and more accurate) other models
1 Qwen3-ASR-1.7Boverall pickmost accuratefast pick 28.2927.34–29.31 3.8× 2.0× 2.04 GB Apache-2.0
2 whisper-large-v3 30.1129.13–31.18 2.1× 0.6× 1.55 GB Apache-2.0
3 whisper-large-v3-turbo 30.5629.52–31.66 3.4× 0.8× 0.83 GB Apache-2.0
4 whisper-medium 42.5741.43–43.78 4.3× 1.1× 0.77 GB Apache-2.0
5 Qwen3-ASR-0.6B 50.3049.30–51.39 8.0× 4.3× 0.79 GB Apache-2.0
6 whisper-small 58.4457.10–59.92 12.1× 3.4× 0.25 GB Apache-2.0