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

Filipino · 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 11.8211.17–12.51 2.1× 0.6× 1.55 GB Apache-2.0
2 whisper-large-v3-turbo 12.0811.43–12.72 3.4× 0.8× 0.83 GB Apache-2.0
3 Fun-ASR-MLT-Nano-2512!overall pickfast pick 15.6214.78–16.63 9.0× 4.5× 0.83 GB FunASR Model Open Source License Agreement v1.1
4 whisper-medium 18.3617.60–19.20 4.3× 1.1× 0.77 GB Apache-2.0
5 Qwen3-ASR-1.7B 24.2923.49–25.18 3.8× 2.0× 2.04 GB Apache-2.0
6 whisper-small 28.5227.52–29.53 12.1× 3.4× 0.25 GB Apache-2.0
7 Qwen3-ASR-0.6B 35.4334.47–36.38 8.0× 4.3× 0.79 GB Apache-2.0

Benchmarked but not listed on the model card: whisper-large-v3, whisper-large-v3-turbo, Fun-ASR-MLT-Nano-2512, whisper-medium, whisper-small.