1 Claude Opus 4.8 Anthropic
88.6% Anthropic — Claude Opus 4.8 vendor unverified2 Claude Opus 4.7 Anthropic
87.6% Anthropic — Claude Opus 4.7 vendor unverified3 Claude Opus 4.6 Anthropic
80.8% Anthropic — Claude Opus 4.6 vendor unverified4 Gemini 3.1 Pro Google DeepMind
80.6% Google DeepMind — Gemini 3.1 Pro model card vendor unverified5 DeepSeek V4-Pro DeepSeek
80.6% DeepSeek — V4-Pro model card vendor unverified6 Qwen3.7 Max Alibaba Qwen
80.4% Qwen — Qwen3.7 Max vendor unverified7 Kimi K2.6 Moonshot AI
80.2% Moonshot — Kimi K2.6 model card vendor unverified8 GPT-5.2 OpenAI
80% llm-stats — GPT-5.2 (vendor-reported) 3rd-party unverified9 Claude Sonnet 4.6 Anthropic
79.6% Anthropic — Claude Sonnet 4.6 vendor unverified10 DeepSeek V4-Flash DeepSeek
79% DeepSeek — V4-Flash model card vendor unverified11 Gemini 3 Flash Google DeepMind
78% Google — Gemini 3 Flash vendor unverified12 Qwen3.6-27B Alibaba Qwen
77.2% Alibaba — Qwen3.6-27B model card vendor unverified13 Gemini 3 Pro Google DeepMind
76.2% Google — Gemini 3 Pro vendor unverified14 Qwen3.6-35B-A3B Alibaba Qwen
73.4% Alibaba — Qwen3.6-35B-A3B model card vendor unverified15 Claude Haiku 4.5 Anthropic
73.3% Anthropic — Claude Haiku 4.5 vendor unverified16 DeepSeek V3.2 DeepSeek
73.1% DeepSeek — V3.2 technical report vendor unverified17 Nemotron 3 Ultra NVIDIA
71.9% NVIDIA — Nemotron 3 Ultra model card vendor unverified18 Kimi K2 Thinking Moonshot AI
71.3% Moonshot — Kimi K2 Thinking model card vendor unverified19 Nova 2 Pro Amazon
61.5% Amazon — Nova 2 technical report vendor unverified20 Gemini 2.5 Pro Google DeepMind
59.6% Google DeepMind — Gemini 2.5 Pro model card vendor unverified