DeepMind researcher's first Kimi 3.1 Pro try goes poorly, with caveats on sample size
SamuelAlbanie · x · 2026-10-25
Samuel Albanie tried Kimi 3.1 Pro on one moderately hard prompt and found the experience wanting — the model spent around two hours reasoning. He caveats that his sample size is one (maybe two hours of reasoning) and the model looks very strong on benchmarks, so he's keen to see wider community reactions.
More from Models
- Self-host open models or risk bans: why 'safety-first' labs make own-hardware the safer bet — gnukeith · 2026-10-25
- CrowdGPT: a crowd-trained, datacenterless 1B open-source LLM, 6.9B tokens done — Vxtzq1 · 2026-10-25
- Miles Brundage asks why Argon isn't available in Google's Antigravity — Miles_Brundage · 2026-10-25
- Apples-to-apples rerun: DeepInfer tops DeepSeek's published throughput by ~9% on 4x B300 — charles_irl · 2026-10-25
- Grok UI shows image icon, hinting at upcoming image upload feature — faraz0x · 2026-10-25
- Polymarket odds put Kimi K4 release below 9% by end of 2026 — Polymarket · 2026-10-25