Open source #Open models Oct 22, 2026 · Originally published Oct 21

IBM releases Granite Speech 5.0 2B under Apache 2.0, with a 5.06% average English WER and keyword biasing for up to 2,000 terms

One update covering accurate transcription, fast on-device transcription and multi-speaker meetings. What each model is for, and how far Chinese support goes.

Primary source Hugging Face Blog · IBM Granite · huggingface.co Read the original ↗

// Key points

granite-speech-5.0-2b transcribes English, French, German, Spanish, Portuguese and Japanese, translates between those and English, and also translates English to Italian and Mandarin; trained on 390,000 hours of public audio, it ranks #2 among open-weight models on the Open ASR Leaderboard at a 5.06% average English short-form WER.
It supports keyword lists of up to 2,000 terms, naming the input language in the prompt, and punctuation and capitalization on request, with support in Transformers, vLLM, llama.cpp and mlx-audio.
The 470M Turbo CTC now covers 7 languages with native punctuation, and its RTFx rose from about 12,600 to 19,870, roughly 1.6x faster.
granite-speech-4.1-2b-plus, for meetings and calls, is a drop-in replacement: long-form WER falls from 13.53 to 10.22, biased-keyword WER with 20-term lists from 11.79 to 6.48, and average timestamp shift from 210.86 to 98.54.

Builder's takeNote that it doesn't transcribe Chinese; Mandarin is only a translation target from English, so Chinese interviews still need another option. For English mock interviews it's worth a try: put the role's technical terms in the keyword list to catch the words that get mangled most, like Kubernetes or gRPC, and use 4.1-2b-plus when you need speaker-split recordings. Run your own recordings through it and Mistral's Voxtral 3 from October 20 before choosing.

// Background · from #Open models

Full timeline →