โ† ์ตœ์‹  ํ”„๋กœ๋•ํŠธ

Getting GLM 5.2 running on my slow computer

Run frontier MoE models on hardware you already own โ€” pure C, zero deps, experts streamed from disk. Tiny engine, immense model. ๐Ÿฆ. A few days ago I found myself trying out GLM 5.2 and was really positively impressed.

์˜คํ”ˆ์†Œ์Šค

45,420 ์Šคํƒ€ ยท GitHub์™ธ๋ถ€ ์‚ฌ์ดํŠธ

github.com ๋ฐฉ๋ฌธ์™ธ๋ถ€ ์‚ฌ์ดํŠธ

๋ฐ˜์‘ (1)

๋น„์Šทํ•œ ํ”„๋กœ๋•ํŠธ

์„ค๋ช…์ด ๋น„์Šทํ•œ ์ˆœ

  1. RunNburn

    github.com

    Run a 295B Moe from a 98GB GGUF on a 64GB RAM Desktop

    Hacker News11์ 

  2. GhostBox

    ghost.charity

    Borrow a disposable little machine from the Global Free Tier

    Hacker News126์ 

  3. Warp

    github.com

    Run DeepSeek v4.1 Flash with 5 GB of RAM at 3.77 tok/s

    Hacker News16์ 

  4. Ollama

    github.com

    Run LLMs on your Mac

    Hacker News284์ 

  5. HART OS

    github.com

    an open-source AI OS built so frontier AI needs no datacenter

    Hacker News18์ 