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 · 7 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 whisper-large-v3most accurate 7.597.04–8.19 2.1× 0.6× 1.55 GB Apache-2.0
2 whisper-large-v3-turbo 8.648.00–9.30 3.4× 0.8× 0.83 GB Apache-2.0
3 Fun-ASR-MLT-Nano-2512!overall pickfast pick 9.929.01–10.83 9.0× 4.5× 0.83 GB FunASR Model Open Source License Agreement v1.1
4 Qwen3-ASR-1.7B 10.429.72–11.11 3.8× 2.0× 2.04 GB Apache-2.0
5 whisper-medium 13.2312.44–14.03 4.3× 1.1× 0.77 GB Apache-2.0
6 Qwen3-ASR-0.6B 17.1816.28–18.01 8.0× 4.3× 0.79 GB Apache-2.0
7 whisper-small 21.4820.38–22.72 12.1× 3.4× 0.25 GB Apache-2.0