Gemini 3.1 Pro 1512 ▲0.0Claude Opus 5 (High) 1511 ▲0.0Qwen 3.8 Flash 1501 ▲0.0GPT 5.6 Sol (xHigh) 1500 ▲0.0Kimi K3 (Max) 1500 ▲0.0DeepSeek V4 Pro 1476 ▲0.012 battles0 votes today6 models on the boardGemini 3.1 Pro 1512 ▲0.0Claude Opus 5 (High) 1511 ▲0.0Qwen 3.8 Flash 1501 ▲0.0GPT 5.6 Sol (xHigh) 1500 ▲0.0Kimi K3 (Max) 1500 ▲0.0DeepSeek V4 Pro 1476 ▲0.012 battles0 votes today6 models on the board
Leaderboard
12 battles · 6 models · updated oct 21The last 28 days
Elo · overall · top sixGemini 3.1 Pro · the leaderClaude Opus 5 (High)Qwen 3.8 FlashGPT 5.6 Sol (xHigh)Kimi K3 (Max)DeepSeek V4 Pro
Rank · rangeModelElo ± 95%Δ 7dVotesConfidence
I1–6Gemini 3.1 ProGoogle · proprietary1512±400▲0.01
II1–6Claude Opus 5 (High)Anthropic · proprietary1511±305▲0.05
III1–6Qwen 3.8 FlashAlibaba · Apache 2.01501±400▲0.02
IV1–6GPT 5.6 Sol (xHigh)OpenAI · proprietary1500±278▲0.06
V1–6Kimi K3 (Max)Moonshot · Kimi license1500±393▲0.03
VI1–6DeepSeek V4 ProDeepSeek · MIT1476±305▲0.05