Self-Introduction
I graduated from the School of EECS, Peking University (22nd Turing Class), also holding a double degree in Economics at the National School of Development.
Currently a full-time member of the RL team at Kimi, focusing on reasoning & coding.
Research interests: RL, post-train, agent.
Selected work
- 🔭 K3 — Core contributor to post-training, with a focus on Knowledge-Graph-Guided Task Synthesis.
- SOTA on AA-LCR; national SOTA on SciCode, CritPt, GPQA, HLE and some coding benchmarks.
Benchmark Kimi K3 Claude Fable 5 GPT-5.6 Sol Claude Opus 4.8 GPQA-Diamond 93.5 92.6 94.1 91.0 CritPt 23.4 28.6 32.3 20.9 AA-LCR 74.7 70.0 73.7 67.7 SciCode 58.7 60.2 56.1 53.5 HLE-Full (w/o tools) 43.5 53.3 44.5 49.8 HLE-Full (w/ tools) 56.0 63.0 58.0 57.9 - 🔭 K2.6
-
🔭 K2.5 — my domain: HLE / AIME / IMO / …; we are the SOTA of HLE !
Benchmark Kimi K2.5 Claude Opus 4.5 GPT-5.2 (xhigh) Gemini 3 Pro DeepSeek-V3.2 HLE-Full (w/o tools) 30.1 30.8 34.5 37.5 25.1 HLE-Full (w/ tools) 50.2 43.2 45.5 45.8 40.8 AIME 2025 96.1 92.8 100 95.0 93.1 IMO-AnswerBench 81.8 78.5 86.3 83.1 78.3 GPQA-Diamond 87.6 87.0 92.4 91.9 82.4 - 🔭 Kimi k2 thinking
- 🚀 PKUHUB — founder & top of the leaderboard (2025-2026)
- Over 30 course notes and 2K stars by Peking University faculty and students
- 🚀 AutoPku — end-to-end PKU academic assistant powered by an expert agent team
- 🥈 2020 CMO (Chinese Mathematical Olympiad) — Silver Medal
