whatmodel.sh / stt / Phonon-2

Phonon-2

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Phonon-2 is an open-weights speech-to-text model from Fermion Research, released September 2026. Its published size is 0.6B parameters. Licence: cc-by-4.0.

Catalogue identity notes

Public checkpoint with maker-documented runtime. Memory is estimated from parameters and precision, not measured; no API price verified in this pass.

Model information sources

Release date ↗

56.1±6.5#62 of 163 speech-to-text modelsfrom 1 benchmark result across 1 source* thinly tested, treat the rank as provisional

Score within the Open ASR English short-form WER (March 2026) comparison set only.

$ how it compares

The speech-to-text models ranked either side of Phonon-2, out of 178 with a score, then more from Fermion Research.

#modelscorewhereRAMreleased
71Parakeet Redux Moondream57±3local~0.2GB2026-09
72Zipformer Transducer XL 290M SoundsGood AI56±5comparison only2026-05
73Phonon-2 Fermion Research56±7local~0.7GB2026-09
74Parakeet TDT-CTC 1.1B NVIDIA56±4local~1.3GB2024-05
75GPT-4o Mini Transcribe OpenAI56±2api $0.0030/audio min2025-03
more from Fermion Research
96Phonon-1 Fermion Research53±7comparison only2026-08

see all speech-to-text models · find the ones that fit your machine

$ what it's good at

other tests1 not in a task · 84% of its score
testresultvs the others in its rundifficultycountswhy
Open ASR seven English sets (Fermion full-set runs) lab5.2% ✎█······█···█ beats 2/2avg 8.4%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 84%small run

$ where to run it

Other ways to run it

  1. Run it as published · needs ~0.7 GB, Mac only · from huggingface.co/FermionResearch (official)
    python3.12 -m venv .venv && source .venv/bin/activate && pip install fermion-research mlx mlx-audio mlx-lm soundfile scipy zstandard && phonon transcribe recording.wav

Official weights and documentation: huggingface.co/FermionResearch ↗

$ how the score was worked out · Phonon-2 model card 84%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
Phonon-2 model card · lab-reported1cc-by-4.0
prior · pull toward the middle
weighted average= 56.1

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json