We’re releasing the Core ML package for Phonon-2, reaching a...
Manan Gupta@yoitsmanan
11 views
Oct 06, 2026
~1 min read
Advertisement
1
We’re releasing the Core ML package for Phonon-2, reaching a staggering 606× real time, up from our 174× launch figure.
That’s an hour of speech in just 6(!) seconds, on a base M5 MacBook Air, with no loss in quality.
It’s now the default in Detta. We’ve cut peak app memory from 3.1 GB to 1.1 GB, and idle memory from 1.8 GB to just 600 MB, even with Gluon loaded!
That puts Detta's peak memory footprint same as WisprFlow, with the models in Detta running completely locally!
That’s an hour of speech in just 6(!) seconds, on a base M5 MacBook Air, with no loss in quality.
It’s now the default in Detta. We’ve cut peak app memory from 3.1 GB to 1.1 GB, and idle memory from 1.8 GB to just 600 MB, even with Gluon loaded!
That puts Detta's peak memory footprint same as WisprFlow, with the models in Detta running completely locally!
2
Gluon, our text polishing model, turns messy dictation into clean, formatted text.
Say “Let’s meet on Tuesday, actually no, Wednesday at noon,” and get “Let’s meet onz Wednesday at noon.” It removes the “ums,” repetitions and false starts, adds punctuation, and formats numbers and dates.
Download and update Detta for free today.
fermionresearch.com/products/detta/
Say “Let’s meet on Tuesday, actually no, Wednesday at noon,” and get “Let’s meet onz Wednesday at noon.” It removes the “ums,” repetitions and false starts, adds punctuation, and formats numbers and dates.
Download and update Detta for free today.
fermionresearch.com/products/detta/
3
We worked extremely closely with the Apple Neural Engine to make Detta ridiculously fast, a typical dictation length can be transcribed in a median 11 ms with polish off!
We've also improved Detta's hotword biasing for names and specialized terminology. On our hotword test, custom-term recall jumped from 54% to 79%, with zero false substitutions, and there's a lot more coming here in the next updates.
Read our in depth dive into Detta here.
fermionresearch.com/research/detta…
We've also improved Detta's hotword biasing for names and specialized terminology. On our hotword test, custom-term recall jumped from 54% to 79%, with zero false substitutions, and there's a lot more coming here in the next updates.
Read our in depth dive into Detta here.
fermionresearch.com/research/detta…
4
We are hiring across Engineering and Research teams, if working on the most awesome of problems excites you, shoot us an email at careers@fermionresearch.com with your coolest work.
Related Links:
CoreML Phonon-2: huggingface.co/FermionResearc…
Detta: fermionresearch.com/products/detta/
CoreML Phonon-2 Swift Package:
github.com/fermionresearc…
pip: pypi.org/project/fermio… — pip install fermion-research
Docker (CPU): ghcr.io/fermionresearc… — docker pull ghcr.io/fermionresearc…
Docker (NVIDIA): ghcr.io/fermionresearc… — docker pull ghcr.io/fermionresearc…
GitHub: github.com/fermionresearc…
Research: one eighth the bits: fermionresearch.com/research/one-e…
Site: fermionresearch.com
Join us: fermionresearch.com/careers/
Related Links:
CoreML Phonon-2: huggingface.co/FermionResearc…
Detta: fermionresearch.com/products/detta/
CoreML Phonon-2 Swift Package:
github.com/fermionresearc…
pip: pypi.org/project/fermio… — pip install fermion-research
Docker (CPU): ghcr.io/fermionresearc… — docker pull ghcr.io/fermionresearc…
Docker (NVIDIA): ghcr.io/fermionresearc… — docker pull ghcr.io/fermionresearc…
GitHub: github.com/fermionresearc…
Research: one eighth the bits: fermionresearch.com/research/one-e…
Site: fermionresearch.com
Join us: fermionresearch.com/careers/