Hacker News: Helldezhttps://news.ycombinator.com/user?id=HelldezHacker News RSShttps://hnrss.org/hnrss v2.1.1Wed, 23 Sep 2026 01:10:42 +0000<![CDATA[Qwen 3.7 or 3.8 30B – 100B. QWhen?]]>I'm not quite following the news.

Is it true?

When will the Qwen versions of this size be released?

Comments URL: https://news.ycombinator.com/item?id=49088705

Points: 6

# Comments: 0

]]>Tue, 28 Jul 2026 19:25:50 +0000https://news.ycombinator.com/item?id=49088705Helldezhttps://news.ycombinator.com/item?id=49088705https://news.ycombinator.com/item?id=49088705<![CDATA[New comment by Helldez in "Qwen 3.8"]]>I tried it and it's incredible

]]>Mon, 20 Jul 2026 19:08:31 +0000https://news.ycombinator.com/item?id=48983455Helldezhttps://news.ycombinator.com/item?id=48983455https://news.ycombinator.com/item?id=48983455<![CDATA[New comment by Helldez in "Show HN: I ported llama.cpp to Apple Watch and ran a 0.8B LLM locally"]]>How?

]]>Mon, 20 Jul 2026 19:04:38 +0000https://news.ycombinator.com/item?id=48983400Helldezhttps://news.ycombinator.com/item?id=48983400https://news.ycombinator.com/item?id=48983400<![CDATA[Show HN: Run a 120B-parameter MoE on Android mid-range phone CPU-only llama.cpp]]>Article URL: https://github.com/Helldez/BigMoeOnEdge

Comments URL: https://news.ycombinator.com/item?id=48969917

Points: 1

# Comments: 0

]]>Sun, 19 Jul 2026 17:17:43 +0000https://github.com/Helldez/BigMoeOnEdgeHelldezhttps://news.ycombinator.com/item?id=48969917https://news.ycombinator.com/item?id=48969917<![CDATA[New comment by Helldez in "Show HN: Biscotti – on-device meeting transcription for macOS"]]>Have you ever considered Parakeet for transcription? It’s lightweight and multilingual

]]>Sun, 19 Jul 2026 15:04:39 +0000https://news.ycombinator.com/item?id=48968800Helldezhttps://news.ycombinator.com/item?id=48968800https://news.ycombinator.com/item?id=48968800