Fanqing(@FanqingMengAI):New RSIBench-Data (https://t.co/LmxqLvvEAk) result: Kimi K3 + Kimi Code achieves a 27.317% weighted score (aime,gpqa use 0.1, others 0.2) across six benchmarks, ranking #1 among the five model–harness combinations we tested. It scores 50% on SWE-bench Verified and 17% on SWE-bench Pro. A strong result for automated RSI research. Congratulate!

Fanqing

@FanqingMengAI

Co-founder of @Evolvent_AI, Previous: @Kimi_Moonshot (focus on rl infra) , Now focus on agentic benchmark & infra, contact: mengfanqing33@gmail.com

가입 March 2025

775 팔로잉 중 2.1K 팬

Fanqing@FanqingMengAI

2026.08.03 02:12

New RSIBench-Data ( result: Kimi K3 + Kimi Code achieves a 27.317% weighted score (aime,gpqa use 0.1, others 0.2) across six benchmarks, ranking #1# among the five model–harness combinations we tested. It scores 50% on SWE-bench Verified and 17% on SWE-bench Pro. A strong result for automated RSI research. Congratulate!