{"id":"phonon-gradium","slug":"phonon-gradium","name":"Phonon","orgId":"gradium","family":"phonon-gradium","releaseDate":null,"availability":"announced","openWeights":true,"license":null,"paramsTotal":null,"paramsSource":null,"contextWindow":null,"knowledgeCutoff":null,"modelCardUrl":"https://gradium.ai/phonon","hfRepo":null,"modalities":["audio"],"aliases":["Phonon","gradium phonon","phonon 1.0","phonon 31m","phonon-31m","phonon-31m-en","gradium/phonon-31m-en","phonon 40m","phonon-40m","phonon 40m multilingual","phonon-40m-multilingual","gradium/phonon-40m-multilingual","gradium/phonon-40m-en","gradium/phonon-40m-fr","gradium/phonon-40m-de","gradium/phonon-40m-es","gradium/phonon-40m-pt"],"aliasNote":null,"origin":"tracked","tracked":{"trackedSince":"2026-10-20","announced":"2026-10-20","blurb":"Gradium's suite of on-device text-to-speech models under 50M parameters, built to run offline on phones, laptops, small boards and in the browser. Phonon 31m (English) is Apache 2.0; the 40m models are CC BY-NC 4.0. Not an LLM.","api":"none","openWeights":"yes","params":{"text":"31M and 40M (excluding the audio tokenizer)","source":"Gradium's Hugging Face model cards and Phonon page"},"facts":[{"label":"Lab","value":"Gradium"},{"label":"Models","value":"Phonon 31m (English), Phonon 40m (one model each for English, French, German, Spanish and Portuguese) and Phonon 40m multilingual (all five)"},{"label":"Weights","value":"Hugging Face: phonon-31m-en, phonon-40m-en/-fr/-de/-es/-pt, phonon-40m-multilingual, plus phonon-voices; q8 files 64.5 MB (31m) and 74.6 MB (40m)"},{"label":"License","value":"Phonon 31m Apache 2.0; all 40m models CC BY-NC 4.0 (commercial use through Gradium sales); phonon-voices Apache 2.0"},{"label":"Builds on","value":"Kyutai's Pocket TTS and the Mimi codec (model card)"},{"label":"Runs on","value":"CPU, Apple Neural Engine (Core ML), CUDA, Metal, Vulkan, Snapdragon NPU, and the browser via WebAssembly / WebGPU (lab statement)"},{"label":"Hugging Face repos created","value":"Oct 19, 2026 (UTC); announced Oct 20, 2026"}],"claims":["Lab claim: 1.70% word error rate for Phonon 31m on 500 hard English sentences (phone numbers, email addresses, postal codes), against 3.50% for Kokoro, 7.40% for Kitten TTS and 8.71% for NeuTTS Nano."],"sources":[{"kind":"lab","title":"Phonon (Gradium product page)","url":"https://gradium.ai/phonon","date":"2026-10-20"},{"kind":"repo","title":"phonon-31m-en on Hugging Face","url":"https://huggingface.co/gradium/phonon-31m-en","date":"2026-10-19"},{"kind":"repo","title":"phonon-40m-multilingual on Hugging Face","url":"https://huggingface.co/gradium/phonon-40m-multilingual","date":"2026-10-19"},{"kind":"repo","title":"phonon-voices on Hugging Face","url":"https://huggingface.co/gradium/phonon-voices","date":"2026-10-19"}],"firstBenchmark":null},"organization":{"id":"gradium","name":"Gradium","slug":"gradium","country":null,"url":"https://gradium.ai","color":"#11110F","mark":"G"},"price":null,"answer":{"word":"Pending benchmarks.","scope":"Announced Oct 20, 2026; no independent results yet.","lead":"As of Oct 22, 2026, Phonon is pending benchmarks on isitsota: announced Oct 20, 2026, with no independent results on the boards we track yet.","confidence":"low"},"status":"PENDING_BENCHMARKS","verdicts":[]}