Gradium
3 models across 1 family · Latest: Phonon 31M (2026-10)
Real-time voice AI: text-to-speech, speech-to-text and live translation.
Gradium's portfolio covers 3 active models across 1 current family, spanning general LLM work. Open a model detail page to compare provider routes and sourced benchmarks.
Covers 0 workload areas across 3 active tracked models; last verified 2026-10-20.
Use it for
- Teams evaluating general LLM work across this lab's releases
- Comparing model families before committing to a flagship
- Migration and pricing follow-ups across 3 tracked models
Do not use it for
- Choosing a hosting provider without opening a model page for price ladders
Active models
3
Current models from this lab, excluding deprecated ones
Active families
1
Current model families from this lab
Open catalog
3 open
1 open source / 2 open weights
Lowest output price
Not tracked
No provider output pricing linked yet
Latest dated release
2026-10-20
Phonon 31M
Freshness
2026-10-20
Researched 1d ago
Information
Release cadence
- Dated releases
- 3
- Latest model
- Phonon 31M
- Latest date
- 2026-10-20
Where this lab wins
Not enough capability or benchmark coverage yet to call strengths for this lab.
Flagship quality / price signal
- Flagship
- Phonon 31M
- Selection source
- Best sourced coding Q/$
- Coding grade
- unknown
- Benchmark / output price
- Not enough sourced benchmark and price coverage
About
Gradium is a Paris-based voice AI company (Gradium SAS, Paris trade register 989 284 955) that builds real-time text-to-speech, speech-to-text, live translation, voice cloning and voice design models, served through its API, and the open on-device Phonon TTS models.
Featured models
| Model | Released | Context | Input price ($/1M) | Output price ($/1M) | License | Openness |
|---|---|---|---|---|---|---|
| Phonon 31M | 2026-10-20 | - | - | - | Apache 2.0 | Open source |
| Phonon 40M | 2026-10-20 | - | - | - | CC-BY-NC-4.0 | Open weights |
| Phonon 40M Multilingual | 2026-10-20 | - | - | - | CC-BY-NC-4.0 | Open weights |
Model families
Recent releases
- Phonon 31M- 2026-10-20
- Phonon 40M- 2026-10-20
- Phonon 40M Multilingual- 2026-10-20
Explore related pages
Last reviewed: 2026-10-20. Data sourced from public lab announcements and provider documentation.