Benchgen

CharXiv Reasoning — Results

RankModelScore
1kimi-k391.3
2fugu-ultra86.6
3fugu85.1
4inkling82
5muse-glimmer78.8
6kimi-k30.913
7claude-opus-4-70.91
8claude-opus-4-80.899
9muse-spark0.884
10claude-sonnet-50.883
11kimi-k2-60.867
12seed-2-1-pro0.864
13qwen3-7-plus0.859
14seed-2-1-turbo0.836
15gpt-5-20.821
16gpt-5-50.816
17qwen3-6-plus0.815
18gemini-3-pro0.814
19gpt-50.811
20gemini-3-flash0.803
21qwen3-5-27b0.795
22o30.786
23qwen3-6-27b0.784
24qwen3-6-35b-a3b0.78
25kimi-k2-50.775

CharXiv Reasoning — Phases

1 phaseActive

Phase 1·

CharXiv Reasoning

Jun 22, 2026
Current