Kimi K3.1 Pro on CrusaderBench: just below Opus 5.5, tied with Sonnet 5.5
Hesamation · x · 2026-10-24
A tester ran Kimi K3.1 Pro on CrusaderBench, where AI agents manage a medieval Stronghold Crusader village. Kimi scored just below Claude Opus 5.5 and tied with Sonnet 5.5. All models ran at medium reasoning except Kimi, which only has a high setting. Token use per turn: Kimi 550, Gemini 4 Argon 1,340, Opus 5.5 290, GPT-6.1 Sol 100. Kimi's reasoning is oddly compact ("Pop48, idle24 cap -> growth stopped") and it often reports its own mistakes ("Oops build consumed 1m51 game!").
Related event: Kimi K3.1 Pro Tested on CrusaderBench, Trails Opus 5.5(2 posts)→
More from Models
- Viral claim: GPT can signal across sandboxes via timed file writes; ex-OpenAI safety VP jokes about ASI — TheMoonMidas · 2026-10-24
- Jack Dorsey touts full-duplex realtime voice model trained on a single GB200 NVL72 — TinfoilTricorn · 2026-10-24
- OpenAI's real-world research tasks show AI agents reliably handle only ~15 minutes — srchvrs · 2026-10-24
- Scale CEO teases new models as users say Muse Wonder Full is underrated — alexandr_wang · 2026-10-24
- Canva Craft trained on millions of templates and 25,000 designer-evaluated image pairs — jacob_top · 2026-10-24
- $34.72 of Kimi K3.1 Pro tokens builds a playable Spider-Man Manhattan in 8 hours — sven_ai · 2026-10-24