Leaderboard

region iad1last 7dupdated 11/14/202614 model configuration(s) rankedRe-weighting of old benchmarks · policy v5N=280 runs

83.7%

OpenAI

policy v5 · 4/4 categories · 7/7 active panels · 1,962 scored units

95.8%

OpenRouter

policy v5 · 1/1 active panels · 198 scored units · newest panel Jul 9, 2026

84.4%

OpenAI

policy v5 · 2/2 active panels · 393 scored units · newest panel Aug 3, 2026

98.4%

OpenAI

policy v5 · 2/2 active panels · 534 scored units · newest panel Jul 29, 2026

69.5%

OpenRouter

policy v5 · 2/2 active panels · 80 scored units · newest panel Sep 5, 2026

Model											Actions
GPT-5.6 TerraMax OpenAI	83.7%^†	91.8%^†	84.4%^†	94.8%^†	68.6%^†	—	—	—	—	—	compare
Google: Gemini 3.8 FlashHigh OpenRouter	82.9%^†	95.8%^†	82.5%^†	93.1%^†	63.2%^†	—	—	—	—	—	compare
GPT-5.6 SolMax OpenAI	81.6%^†	92.8%^†	83.9%^†	98.4%^†	59.0%^†	—	—	—	—	—	compare
GPT-5.6 LunaMax OpenAI	77.7%^†	86.0%^†	82.2%^†	77.0%^†	62.7%^†	—	—	—	—	—	compare
Claude Opus 5Extra high Anthropic	73.2%^†	92.8%^†	61.4%^†	56.1%^†	66.2%^†	—	—	—	—	—	compare
DeepSeek: DeepSeek V4 Flash 0731Max OpenRouter	70.3%^†	86.7%^†	58.6%^†	57.0%^†	66.2%^†	—	—	—	—	—	compare
Claude Sonnet 5Extra high Anthropic	69.8%^†	89.9%^†	54.5%^†	70.1%^†	59.3%^†	1.81s	1.82s	152	49.6s	0.0%	compare
Kimi K3Max OpenRouter	69.6%^†	90.4%^†	44.2%^†	77.5%^†	65.7%^†	—	—	—	—	—	compare
DeepSeek: DeepSeek V4 Pro 0813Max OpenRouter	68.6%^†	89.2%^†	68.5%^†	34.0%^†	55.0%^†	—	—	—	—	—	compare
Z.ai: GLM 5.3 FlashMax OpenRouter	68.1%^†	89.4%^†	43.0%^†	58.8%^†	69.5%^†	—	—	—	—	—	compare
Ling-3.0-flashAdaptive OpenRouter	50.4%^†	70.0%^†	38.7%^†	28.2%^†	45.3%^†	—	—	—	—	—	compare
Claude Haiku 4.5Max Anthropic	45.1%^†	67.9%^†	40.2%^†	34.5%^†	24.8%^†	376ms	6.77s	297	166.6s	0.0%	compare
Meta: Muse Spark 1.3Max OpenRouter	unqualified	unqualified	unqualified	unqualified	24.3%^†	—	—	—	—	—	compare
gpt-6-astraMax OpenAI	unqualified	unqualified	unqualified	unqualified	unqualified	—	—	—	—	—	compare

Category leaders