Hacker News: mysterhawk

New comment by mysterhawk in "Ask HN: Who wants to be hired? (May 2026)"

mysterhawk — Wed, 06 May 2026 12:00:20 +0000

Location: Italy

Remote: Yes

Willing to relocate: No

Technologies: Rust,Python,Java-spring,TypeScript,React,Html,CSS,Nim,C,C++,PHP

Resumé/CV: https://drive.google.com/file/d/1F_Fp3pKusTQtVBgcxYF6h_2leOz... or https://www.linkedin.com/in/antonio-gagliarducci

Email: gagliarducci.antonio@posteo.net

New comment by mysterhawk in "Serving 70B-Scale LLMs Efficiently on Low-Resource Edge Devices [pdf]"

mysterhawk — Thu, 03 Oct 2024 16:09:03 +0000

With this model:

https://huggingface.co/bartowski/Qwen2.5-72B-Instruct-GGUF

Man I need to test the q8 version with llamafiles optimizations, it would be so nice to host it locally with the new ryzens, it could maybe fit my 96GB of ram

New comment by mysterhawk in "Serving 70B-scale LLMs efficiently on low-resource edge devices [pdf]"

mysterhawk — Thu, 03 Oct 2024 16:06:20 +0000

https://justine.lol/matmul/

https://www.youtube.com/watch?v=-mRi-B3t6fA

Checkout these articles

New comment by mysterhawk in "Serving 70B-scale LLMs efficiently on low-resource edge devices [pdf]"

mysterhawk — Thu, 03 Oct 2024 16:05:16 +0000

I could run at 1token/s on qwen 2.5 72b q4_k_m on my i7 8750h + alot of ram XD

New comment by mysterhawk in "Serving 70B-Scale LLMs Efficiently on Low-Resource Edge Devices [pdf]"

mysterhawk — Thu, 03 Oct 2024 16:04:24 +0000

And everything runs way faster on cpu like that

New comment by mysterhawk in "Serving 70B-scale LLMs efficiently on low-resource edge devices [pdf]"

mysterhawk — Thu, 03 Oct 2024 16:04:08 +0000

You can already do it with llamafile, checkout the project, it lets you convert a .gguf model in a portable executable