byeongsoo kang
Author ProfileClaim This Author Profile
Prove ownership by publishing #HashtagPLUS and this profile link on your author page or an article under your byline. A moderator or admin will review the request before it merges into your real HashtagPLUS username.
🌐 dev.toSource
I pulled the just-released Gemma 4 12B and ran it on a GTX 1080 Ti. ~28 tok/s at Q4 on one card — but three things broke first, and Q8 (2 cards, 30% slower) fixed both the token glitches and a domain answer Q4 got confidently wrong.
🌐 dev.toSource
I benchmarked Qwen3.6-35B-A3B (IQ4_XS) on two 8-year-old GTX 1080 Ti cards: ~20 tok/s, and the second GPU only adds ~20% (not 2x). Real numbers, VRAM math, and why a 35B fits 22GB.
🌐 dev.toSource
On a small synthetic cohort the graph network scores near-random in leak-free CV (AUC 0.53) — yet the 5-evidence consensus puts known markers in 9 of its top 10. Architecture, real results, and why a weak single model is still a useful voter.