context.vn
Menu

charxiv No Tools

14 models evaluated

#ModelProviderTypeScore
1Claude Mythos 5AnthropicProprietary88.9%
2Qwen3.8 MaxAlibabaOpen weights88.4%
3Gemini 3.8 FlashGoogleProprietary86.2%
4Kimi K3Moonshot AI · Closed84.8%
5Qwen3.8-Flash-NextAlibaba · Open weight84.6%
6Gemini 3.7 FlashGoogle · Closed84.5%
7Qwen3.8-27BAlibaba · Open weight83.7%
8Qwen3.8-Omni-FlashAlibaba · Closed83.5%
9dots3-note PreviewDots Studio · Open weight83.1%
10Claude Opus 4.7 (Adaptive)Anthropic · Closed82.1%
11Claude Opus 4.8Anthropic · Closed80.5%
12InklingThinking Machines Lab · Open weight78.1%
13Inkling-SmallThinking Machines Lab · Open weight77.4%
14Claude Sonnet 5Anthropic · Closed77%