llama.cpp adds Kimi K3.1 Pro support, letting you run Moonshot's giant model locally
jacek2023 · reddit · 2026-11-07
PR #31779 by usrlocalben adds text model support for Kimi K3.1 Pro to llama.cpp (15 files, +919/-35), with weights on Hugging Face under moonshotai/Kimi-K3.1-Pro. The Reddit poster quips "good luck" running it locally, hinting at extreme memory requirements for consumer hardware.
More from Infra
- Copper prices fall as markets fear an AI backlash, reports Financial Times — nordicinst · 2026-11-07
- macOS CI at $0.16/min pays for a Mac Mini in 10 days — why CI prices should drop 50%+ — gajesh · 2026-11-07
- Musk unveils Terafab: a chip fab targeting 1 TW/year of AI compute for Earth and orbit — elonmusk · 2026-11-07
- Musk lays out SpaceX vision of deploying up to 1 TW of space-based AI compute per year — XFreeze · 2026-11-07
- Darkbloom's Mac-based CI network cuts costs 50% and doubles performance — gajesh · 2026-11-07
- Groq kicks off new fundraising at $8 billion pre-money valuation — nmasc_ · 2026-11-07