Link · kept 6 February 2026
Kimi 2.5 is as good as Sonnet 4.5, ChatGPT O3 and Grok 4
Source XIt’s crazy how good good open source LLMs have gotten.
Kimi K2.5 set a new record among open-weight models on the Epoch Capabilities Index (ECI), which combines multiple benchmarks onto a single scale. Its score of 147 is about on par with o3, Grok 4, and Sonnet 4.5. It still lags the overall frontier.
❦
☞ Threads from here
Nearby leaves, connected by subject and form.
234 My adventures in vibe coding shares ai · both link 228 Age of anxiety shares ai · both link 238 Big tech capex numbers are ridiculous shares ai · both link 240 Predictions are hard shares ai · both link 225 Psychoanalysing Dario Amodei shares ai · both link 244 Things we lost in the fire shares ai · both link
browse all subjects →