Hacker GitHub

← All repos

brontoguana/krasis

View on GitHub ↗ ★ 528 C++

Krasis is a Hybrid LLM runtime which focuses on efficient running of larger models on consumer grade VRAM limited hardware

4 mentions on Hacker News by 3 people: 1 story, 3 comments · first 2026-03-17 · latest 2026-06-18

Mentions per year 4 mentions · busiest year 2026 (4)
20082026
AllStoriesComments NewestTopOldest
  1. …or 35-A3B is "near-Opus level" Uh, so, yeah. Im running local Qwen, but Qwen3.5-122B using Krasis https://github.com/brontoguana/krasis Its by far better than Opus. In fact with a phone migration, I was using an OLD android 2fa app…
    comment by mystraline · 2026-06-18 · in Local Qwen isn't a worse Opus, it's a different tool
  2. Check out Krasis https://github.com/brontoguana/krasis It enables something similar to unified memory. Ive got a 5060 (16GB) card and 96 GB ddr5. I can run qwen3.5-122b int4 at 25tok/sec.And now even does image ingestion! Ive been bulk…
    comment by mystraline · 2026-06-09 · in Ask HN: What are tools you have made for yourself since the advent of AI?
  3. https://github.com/brontoguana/krasis On my desktop RTX 5060 TI (16GB) and 96GB ram, I routinely get 25-30 tokens/sec using an 80B model quantized to int8. Uses 65GB system ram and 15GB gfx ram. And its plenty fast for many of my purposes.…
    comment by nekusar · 2026-04-04 · in Tell HN: Anthropic no longer allowing Claude Code subscriptions to use OpenClaw