Hacker News: febed

New comment by febed in "TinySol, a tiny solitaire game for DOS"

febed — Sun, 09 Aug 2026 02:03:01 +0000

8086 and up! Amazing

New comment by febed in "Show HN: Open-source engine running Gemma 4 26B in 2 GB RAM on any M-series Mac"

febed — Fri, 31 Jul 2026 15:35:46 +0000

Thanks a lot for the recommendations!

New comment by febed in "The AI trade now runs on borrowed money, and the lenders are repricing it"

febed — Fri, 31 Jul 2026 06:38:29 +0000

The link is broken

New comment by febed in "Show HN: Open-source engine running Gemma 4 26B in 2 GB RAM on any M-series Mac"

febed — Fri, 31 Jul 2026 04:26:06 +0000

Is the model response quality identical to the memory unconstrained model?

New comment by febed in "Show HN: Open-source engine running Gemma 4 26B in 2 GB RAM on any M-series Mac"

febed — Fri, 31 Jul 2026 04:19:22 +0000

Which model do you think currently hits the highest size-to-performance ratio for agentic tasks like tool calling?

New comment by febed in "Show HN: Open-source engine running Gemma 4 26B in 2 GB RAM on any M-series Mac"

febed — Fri, 31 Jul 2026 04:08:08 +0000

Curious if the same idea could work with gpt-oss-120b? So one could run at least slowly on a Mac

New comment by febed in "Show HN: Open-source engine running Gemma 4 26B in 2 GB RAM on any M-series Mac"

febed — Thu, 30 Jul 2026 03:09:39 +0000

Can the same be done with qwen3.6-35b-a3b?

New comment by febed in "Nativ: Run frontier open models locally on your Mac"

febed — Wed, 22 Jul 2026 04:00:14 +0000

Interesting project. So I guess the search tool is to crawl Wikipedia for articles? And how do you ensure that the tagging stays within the Wikidata taxonomy? How exactly are you using a geocoder? Sorry, just curious

New comment by febed in "Nativ: Run frontier open models locally on your Mac"

febed — Tue, 21 Jul 2026 06:19:16 +0000

What exactly do you mean by data cleaning

New comment by febed in "Vāgdhenu: A Sanskrit Chanting TTS System"

febed — Sat, 18 Jul 2026 08:44:11 +0000

Would be interesting to make a similar one for Tibetan chants.

New comment by febed in "Migrating a production AI agent to GPT-5.6: 2.2x faster, 27% cheaper"

febed — Mon, 13 Jul 2026 03:59:55 +0000

What SDK are you using? Or is it custom?

New comment by febed in "I Built a Telegram Client for Pi"

febed — Wed, 08 Jul 2026 20:16:53 +0000

Off topic, but apart from pi-tui, is there a recommended TUI library that integrates well with the Pi? I want to have a multi pane TUI experience like lazydocker right inside Pi. Pi-tui is a bit limited.

New comment by febed in "Ternlight – 7 MB embedding model that runs in browser (WASM)"

febed — Wed, 08 Jul 2026 20:02:10 +0000

Apart from maximizing compatibility, is there a technical reason why GPU is not used if available?

New comment by febed in "Ask HN: What are you building first with Fable?"

febed — Thu, 02 Jul 2026 04:30:58 +0000

I’d like to build a custom harness - but would that get you banned?

Society of Saint Pius X

febed — Thu, 02 Jul 2026 04:27:16 +0000

Article URL: https://en.wikipedia.org/wiki/Society_of_Saint_Pius_X

Comments URL: https://news.ycombinator.com/item?id=48756567

Points: 25

# Comments: 0

OMLX v0.3.9 Stable Merges Native MTP (Multi-Token Prediction)

febed — Fri, 22 May 2026 06:42:03 +0000

Article URL: https://github.com/jundot/omlx/releases/tag/v0.3.9

Comments URL: https://news.ycombinator.com/item?id=48232765

Points: 2

# Comments: 0

New comment by febed in "Remove–AI–Watermarks – CLI and library for removing AI watermarks from images"

febed — Wed, 20 May 2026 08:02:57 +0000

Since it seems like you would know, could it also work with Non AI static watermarks like repeating oblique lines?

New comment by febed in "Go on Embedded Systems and WebAssembly"

febed — Sat, 04 Apr 2026 05:04:03 +0000

Does it support ESP32

New comment by febed in "Show HN: Gemini can now natively embed video, so I built sub-second video search"

febed — Wed, 25 Mar 2026 04:47:37 +0000

This seems like something that would be very expensive to run. Do you have some representative figures at a particular resolution and frame rate?

New comment by febed in "Show HN: Context Mode – 315 KB of MCP output becomes 5.4 KB in Claude Code"

febed — Sun, 01 Mar 2026 05:44:11 +0000

It would be helpful if you could add a diagram showing the flow of data. I get the general idea but not how you implemented it.