The AI arena is free today

Open Superagent
AI OrganizationMaintains 7 benchmarks

Peking University: AI benchmarks & research

Peking University research groups behind Claw-Eval, PHYBench, Video SimpleQA, MLVU and STEM.

Maintained by Peking University

7 benchmarks

In one line

Peking University maintains Claw-Eval (counts toward the LLM Stats Score, 16 models scored), ClawEval-MM (counts toward the LLM Stats Score, 6 models scored), MLVU (counts toward the LLM Stats Score, 10 models scored), MLVU-M (counts toward the LLM Stats Score, 9 models scored), PHYBench (counts toward the LLM Stats Score, 3 models scored), STEM (counts toward the LLM Stats Score, 1 model scored) and VideoSimpleQA (counts toward the LLM Stats Score, 4 models scored). It ships no models of its own on LLM Stats.

FAQ

Common questions about Peking University.

What is Peking University?

Peking University research groups behind Claw-Eval, PHYBench, Video SimpleQA, MLVU and STEM. On LLM Stats it maintains 7 benchmarks, led by Claw-Eval, MLVU and MLVU-M. It ships no models of its own; each score on its benchmarks carries its own provenance label on the model page.

Which Peking University benchmarks count toward the LLM Stats Score?

Claw-Eval, MLVU, MLVU-M, ClawEval-MM, VideoSimpleQA, PHYBench and 1 more count toward the LLM Stats Score. Only benchmarks admitted to the core set move the main leaderboard; the rest are shown for transparency.

Which models lead Peking University's benchmarks?

Kimi K2.6 leads Claw-Eval at 80.9% (16 models scored), Qwen3.7-Plus leads MLVU at 87.4% (10 models scored) and Qwen3 VL 32B Instruct leads MLVU-M at 82.1% (9 models scored). Each benchmark page lists the full ranking and how the results were produced.

Where are Peking University's official benchmark results?

Peking University's official site is https://www.pku.edu.cn. Each benchmark page on LLM Stats links to that benchmark's own paper or leaderboard, and shows the scores next to every other benchmark so models can be compared on one page.