Hacker News: abhikul0

New comment by abhikul0 in "A True Life Hack: What Physical 'Life Force' Turns Biology's Wheels?"

abhikul0 — Thu, 23 Apr 2026 06:32:24 +0000

Relevant Smarter Every Day video: https://www.youtube.com/watch?v=VPSm9gJkPxU

New comment by abhikul0 in "Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All"

abhikul0 — Thu, 16 Apr 2026 15:09:32 +0000

I'll try to use that, but llama-server has mmap on by default and the model still takes up the size of the model in RAM, not sure what's going on.

New comment by abhikul0 in "Qwen3.6-35B-A3B: Agentic coding power, now open to all"

abhikul0 — Thu, 16 Apr 2026 14:39:38 +0000

Of course the swap is there for fallback but I hate using it lol as I don't want to degrade SSD longevity.

New comment by abhikul0 in "Qwen3.6-35B-A3B: Agentic coding power, now open to all"

abhikul0 — Thu, 16 Apr 2026 14:36:13 +0000

For the 9B model, I can use the full context with Q8_0 KV. This uses around ~16GB, while still leaving a comfortable headroom.

Output after I exit the llama-server command:

  llama_memory_breakdown_print: | memory breakdown [MiB]  | total    free     self   model   context   compute    unaccounted |
  llama_memory_breakdown_print: |   - MTL0 (Apple M3 Pro) | 28753 = 14607 + (14145 =  6262 +    4553 +    3329) +           0 |
  llama_memory_breakdown_print: |   - Host                |                   2779 =   666 +       0 +    2112                |

New comment by abhikul0 in "Qwen3.6-35B-A3B: Agentic coding power, now open to all"

abhikul0 — Thu, 16 Apr 2026 14:23:59 +0000

Mac has unified memory, so 36GB is 36GB for everything- gpu,cpu.

New comment by abhikul0 in "Qwen3.6-35B-A3B: Agentic coding power, now open to all"

abhikul0 — Thu, 16 Apr 2026 14:20:02 +0000

A usable quant, Q5_KM imo, takes up ~26GB[0], which leaves around ~6-7GB for context and running other programs which is not much.

[0] https://huggingface.co/unsloth/Qwen3.5-35B-A3B-GGUF?show_fil...

New comment by abhikul0 in "Qwen3.6-35B-A3B: Agentic coding power, now open to all"

abhikul0 — Thu, 16 Apr 2026 14:06:23 +0000

I hope the other sizes are coming too(9B for me). Can't fit much context with this on a 36GB mac.

New comment by abhikul0 in "The local LLM ecosystem doesn’t need Ollama"

abhikul0 — Thu, 16 Apr 2026 10:39:09 +0000

Have you ever tried going to the model registry and seeing that the model was recently updated? What updated? What changed? Should I re-download this 20GB file?

I guess if you're not frustrated with things like this then sure, no need to stop using it.

New comment by abhikul0 in "The local LLM ecosystem doesn’t need Ollama"

abhikul0 — Thu, 16 Apr 2026 10:27:43 +0000

https://en.wikipedia.org/wiki/List_of_generic_and_genericize...

New comment by abhikul0 in "Costasiella kuroshimae"

abhikul0 — Thu, 16 Apr 2026 08:38:39 +0000

Real Science video on the slug: https://www.youtube.com/watch?v=IH_uv4h2xYM

New comment by abhikul0 in "Ask HN: Can't view public pull requests without logging in (GitHub)"

abhikul0 — Wed, 15 Apr 2026 14:53:40 +0000

I thought github was being wonky but yeah, getting 401 Unauthorized error.

Edit: Can't view discussions as well.

New comment by abhikul0 in "God Sleeps in the Minerals"

abhikul0 — Wed, 15 Apr 2026 14:29:48 +0000

And they say: https://www.youtube.com/watch?v=IfpMknlL-pg

New comment by abhikul0 in "In the brain, objects seen and imagined follow the same neural path"

abhikul0 — Tue, 14 Apr 2026 15:41:15 +0000

>Scientists believe people with aphantasia use words or concepts to recall what they've seen.

Like the text-encoder assisted Diffusion models?

New comment by abhikul0 in "Tell HN: Reddit now demands to know why you won't use their app"

abhikul0 — Mon, 13 Apr 2026 06:37:22 +0000

When the app only shows posts that are more than 10 hours old even when sorting by "hot" and shoving down the algorithmic feed on the app's home page, how are people still using the app?

Lately I've only been visiting a few subs that I'm interested in and keeping them open in safari with ublock; it's been a far better experience. This has drastically cut my reddit time now and if I do want to mindlessly scroll, I just use redlib(hosted in docker or one of their public instances)[0]. It has the same "sort" that's used on the desktop site.

[0] https://github.com/redlib-org/redlib

New comment by abhikul0 in "Floating point from scratch: Hard Mode"

abhikul0 — Tue, 07 Apr 2026 15:09:56 +0000

Worked with Retiming using DC Compiler in an ASIC implementation. Remember a lot of back & forth, sometimes the tool just doesn't add enough registers to meet the constraint, had to test variable register depths; this was a design that used Synopsys DesignWare for FP ops lol.

New comment by abhikul0 in "Every GPU That Mattered"

abhikul0 — Tue, 07 Apr 2026 11:30:40 +0000

The 9400 GT mattered to me as it was my first gpu. Had bought NFS Carbon only to find that the home pc only had a CD drive not DVD lol, so finally with that drive upgrade also came the 9400 GT and fun ensued.

New comment by abhikul0 in "Are We Idiocracy Yet?"

abhikul0 — Tue, 07 Apr 2026 11:07:44 +0000

More at https://www.reddit.com/r/tragedeigh/

New comment by abhikul0 in "Google releases Gemma 4 open models"

abhikul0 — Thu, 02 Apr 2026 17:16:45 +0000

Thanks for this release! Any reason why 12B variant was skipped this time? Was looking forward for a competitor to Qwen3.5 9B as it allows for a good agentic flow without taking up a whole lotta vram. I guess E4B is taking its place.

New comment by abhikul0 in "The Windows equivalents of the most used Linux commands"

abhikul0 — Thu, 02 Apr 2026 09:39:48 +0000

Or you can just prepend `wsl` to the linux command you want to run; of course only if you have wsl setup.

https://learn.microsoft.com/en-us/windows/wsl/filesystems#ru...

New comment by abhikul0 in "Tell HN: Litellm 1.82.7 and 1.82.8 on PyPI are compromised"

abhikul0 — Wed, 25 Mar 2026 05:34:57 +0000

Well, I reinstalled and finally upgraded to Tahoe.