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

Indonesian · 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 Qwen3-ASR-1.7Bmost accurate 5.374.85–5.93 3.8× 2.0× 2.04 GB Apache-2.0
2 whisper-large-v3 6.085.48–6.67 2.1× 0.6× 1.55 GB Apache-2.0
3 whisper-large-v3-turbo 7.206.56–7.82 3.4× 0.8× 0.83 GB Apache-2.0
4 Fun-ASR-MLT-Nano-2512!overall pickfast pick 7.526.85–8.23 9.0× 4.5× 0.83 GB FunASR Model Open Source License Agreement v1.1
5 Qwen3-ASR-0.6B 8.497.85–9.21 8.0× 4.3× 0.79 GB Apache-2.0
6 whisper-medium 10.7910.01–11.69 4.3× 1.1× 0.77 GB Apache-2.0
7 whisper-small 18.0217.02–19.15 12.1× 3.4× 0.25 GB Apache-2.0