Standardized EvaluationArtificial Analysis

Frontier AI Leaderboard

Live evaluation metrics across intelligence, reasoning, coding, and economic efficiency. Updated automatically from OpenRouter.

RankModelCreatorIQ IndexCodingAgenticEngineeringContextPrompt / 1MAction
#1Anthropic: Claude Fable 5.2anthropic64.6--72.6977k$8.00Specs
#2Anthropic: Claude Fable 5.2 (batch)anthropic64.6--72.6977k$4.00Specs
#3OpenAI: GPT-6.2 Astraopenai62.1--66.61025k$8.00Specs
#4OpenAI: GPT-6.2 Astra (batch)openai62.1--66.61025k$4.00Specs
#5Google: Gemini 4 Neongoogle61.4--66.81024k$3.00Specs
#6Google: Gemini 4 Neon (batch)google61.4--66.81024k$1.50Specs
#7Google: Gemini 4 Carbongoogle60--65.31024k$2.00Specs
#8Google: Gemini 4 Carbon (batch)google60--65.31024k$1.00Specs
#9Anthropic: Claude Opus 5.5anthropic59.1--65.3977k$4.00Specs
#10Anthropic: Claude Opus 5.5 (batch)anthropic59.1--65.3977k$2.00Specs
#11OpenAI: GPT-6.2 Solopenai57.3--61.21025k$2.00Specs
#12OpenAI: GPT-6.2 Sol (batch)openai57.3--61.21025k$1.00Specs
#13OpenAI: GPT-6 Astraopenai56.9--61.81025k$10.00Specs
#14OpenAI: GPT-6 Astra (batch)openai56.9--61.81025k$5.00Specs
#15OpenAI: GPT-6.1 Solopenai54.6--59.41025k$2.00Specs
#16OpenAI: GPT-6.1 Sol (batch)openai54.6--59.41025k$1.00Specs
#17Anthropic: Claude Sonnet 5.5anthropic54.5--58.5977k$2.00Specs
#18Anthropic: Claude Sonnet 5.5 (batch)anthropic54.5--58.5977k$1.00Specs
#19Google: Gemini 4 Hydrogengoogle53.9--60.11024k$1.50Specs
#20Google: Gemini 4 Hydrogen (batch)google53.9--60.11024k$0.75Specs
#21OpenAI: GPT-6.1 Terraopenai51.3--54.71025k$1.00Specs
#22OpenAI: GPT-6.1 Terra (batch)openai51.3--54.71025k$0.50Specs
#23Google: Gemini 4 Argongoogle50.4--57.31024k$2.00Specs
#24Google: Gemini 4 Argon (batch)google50.4--57.31024k$1.00Specs
#25Anthropic: Claude Fable 5.1anthropic50--57.2977k$10.00Specs
#26Anthropic: Claude Fable 5.1 (batch)anthropic50--57.2977k$5.00Specs
#27MoonshotAI: Kimi K3.1 Promoonshotai46.5--50.91024k$0.24Specs
#28MoonshotAI: Kimi K3.1 Pro (batch)moonshotai46.5--50.91024k$0.37Specs
#29OpenAI: GPT-6 Solopenai44.2--48.91025k$2.00Specs
#30OpenAI: GPT-6 Sol (batch)openai44.2--48.91025k$1.00Specs
#31Anthropic: Claude Opus 5anthropic44.2--51.3977k$5.00Specs
#32Anthropic: Claude Opus 5 (batch)anthropic44.2--51.3977k$2.50Specs
#33Meta: Muse Wondermeta43.6--49.61024k$4.00Specs
#34Qwen: Qwen4 Maxqwen42.9--50.1977k$1.42Specs
#35Meta: Muse Spark 1.3meta36.6--44.11024k$1.25Specs
#36Anthropic: Claude Haiku 5.5anthropic36.5--42.5977k$0.10Specs
#37Anthropic: Claude Haiku 5.5 (batch)anthropic36.5--42.5977k$0.05Specs
#38Meta: Muse Embermeta36--41.81024k$0.10Specs
#39SpaceXSI: Grok 4.7x-ai35.5--41.8488k$2.00Specs
#40SpaceXSI: Grok 4.7 (batch)x-ai35.5--41.8488k$1.60Specs
#41Tencent: Hy4tencent34.6--41.11024k$0.42Specs
#42Qwen: Qwen3.8 Max (0902)qwen33.7--39.3977k$2.00Specs
#43Qwen: Qwen3.8 Max (0902) (batch)qwen33.7--39.3256k$1.15Specs
#44DeepSeek: DeepSeek V4.1 Flash 1017deepseek32.6--37.61024k$0.15Specs
#45DeepSeek: DeepSeek V4.1 Flash 1017 (batch)deepseek32.6--37.61024k$0.04Specs
#46OpenAI: GPT-5.6 Terraopenai32.3--36.61025k$2.00Specs
#47OpenAI: GPT-5.6 Terra (batch)openai32.3--36.61025k$1.00Specs
#48Z.ai: GLM 5.3z-ai31.7--37.61024k$0.07Specs
#49Z.ai: GLM 5.3 (batch)z-ai31.7--37.61024k$0.28Specs
#50SpaceXSI: Grok 4.6x-ai31--37.4488k$2.00Specs