HN user

MysticOracle

134 karma
Posts9
Comments17
View on HN
Gemma 4 on iPhone 4 months ago

Crashes for me on a couple of different iDevices (2 generations behind) after only a few 2-3 chats. Probably not enough RAM.

Saw this one on X the other day updated with Gemma 4 and they have the built-in Apple Foundation model, Qwen3.5, and other models:

Locally AI - https://locallyai.app/

According to Mosyle’s analysis, ModStealer is being delivered to victims through malicious job recruiter ads targeting developers. It uses a heavily obfuscated JavaScript file written with NodeJS that remains completely undetectable by signature-based defenses. And this one isn’t just targeting Mac users either; Windows and Linux environments are also at risk.

New in v3

- Automatic punctuation and capitalization

- Accurate word-level and segment-level timestamps

- Long audio transcription, supporting audio up to 24 minutes long with full attention (on A100 80GB) or up to 3 hours with local attention.

- Released under a permissive CC BY 4.0 license

- Now supports 25 Languages:

Bulgarian (bg), Croatian (hr), Czech (cs), Danish (da), Dutch (nl), English (en), Estonian (et), Finnish (fi), French (fr), German (de), Greek (el), Hungarian (hu), Italian (it), Latvian (lv), Lithuanian (lt), Maltese (mt), Polish (pl), Portuguese (pt), Romanian (ro), Slovak (sk), Slovenian (sl), Spanish (es), Swedish (sv), Russian (ru), Ukrainian (uk)