Get started

Agentic Index

Artificial Analysis's Agentic Index: a composite of the publisher's own agentic and tool-use evaluation runs, on a 0 to 100 scale, as OpenRouter carries it. 94 models ranked, as carried in OpenRouter's catalog.

retrieved sha256 · 0bb53a77713fartificialanalysis.ai ↗

Most capable57.2Claude Fable 5.1Anthropic
Best value613Ling 3.0 Flash19.3 at $0.032 blended per M
Newest ranked28.4Motif 3listed 2 Nov 2026

The top 26 of 94, 57.2 down to 35.2 on the Agentic Index. 7 of them are on the frontier below, in the accent. 0 of the 26 columns match the selection below and are drawn white.

025507510057.2Claude Fable 5.156.5Claude Opus 556.0Qwen3.8 Max (0902)55.5Muse Spark 1.353.1ZGLM 5.352.5Grok 4.651.0GPT-6 Astra50.9ZGLM 5.3 Flash50.7Claude Fable 550.1Qwen3.8 2.4T A95B47.5DeepSeek V4 Flash Vision Exp47.4GPT-5.6 Sol45.8Qwen3.8 27B43.6Claude Sonnet 543.2Muse Spark 1.243.2GPT-5.6 Terra42.1GPT-5.6 Luna41.9Claude Opus 4.841.3DeepSeek V4 Pro 081341.2Grok 4.541.0DeepSeek V4 Flash 073140.2Gemini 3.8 Flash38.6Claude Opus 4.738.4ZGLM 5.236.4GPT-5.535.2Gemini 3.7 FlashAgentic Index, 0 to 100Claude Fable 5.1: 57.2 on the Agentic Index · AnthropicClaude Fable 5.157.2 on the Agentic Index · AnthropicClaude Opus 5: 56.5 on the Agentic Index · AnthropicClaude Opus 556.5 on the Agentic Index · AnthropicQwen3.8 Max (0902): 56.0 on the Agentic Index · QwenQwen3.8 Max (0902)56.0 on the Agentic Index · QwenMuse Spark 1.3: 55.5 on the Agentic Index · MetaMuse Spark 1.355.5 on the Agentic Index · MetaGLM 5.3: 53.1 on the Agentic Index · Z.aiGLM 5.353.1 on the Agentic Index · Z.aiGrok 4.6: 52.5 on the Agentic Index · SpaceXAIGrok 4.652.5 on the Agentic Index · SpaceXAIGPT-6 Astra: 51.0 on the Agentic Index · OpenAIGPT-6 Astra51.0 on the Agentic Index · OpenAIGLM 5.3 Flash: 50.9 on the Agentic Index · Z.aiGLM 5.3 Flash50.9 on the Agentic Index · Z.aiClaude Fable 5: 50.7 on the Agentic Index · AnthropicClaude Fable 550.7 on the Agentic Index · AnthropicQwen3.8 2.4T A95B: 50.1 on the Agentic Index · QwenQwen3.8 2.4T A95B50.1 on the Agentic Index · QwenDeepSeek V4 Flash Vision Exp: 47.5 on the Agentic Index · DeepSeekDeepSeek V4 Flash Vision Exp47.5 on the Agentic Index · DeepSeekGPT-5.6 Sol: 47.4 on the Agentic Index · OpenAIGPT-5.6 Sol47.4 on the Agentic Index · OpenAIQwen3.8 27B: 45.8 on the Agentic Index · QwenQwen3.8 27B45.8 on the Agentic Index · QwenClaude Sonnet 5: 43.6 on the Agentic Index · AnthropicClaude Sonnet 543.6 on the Agentic Index · AnthropicMuse Spark 1.2: 43.2 on the Agentic Index · MetaMuse Spark 1.243.2 on the Agentic Index · MetaGPT-5.6 Terra: 43.2 on the Agentic Index · OpenAIGPT-5.6 Terra43.2 on the Agentic Index · OpenAIGPT-5.6 Luna: 42.1 on the Agentic Index · OpenAIGPT-5.6 Luna42.1 on the Agentic Index · OpenAIClaude Opus 4.8: 41.9 on the Agentic Index · AnthropicClaude Opus 4.841.9 on the Agentic Index · AnthropicDeepSeek V4 Pro 0813: 41.3 on the Agentic Index · DeepSeekDeepSeek V4 Pro 081341.3 on the Agentic Index · DeepSeekGrok 4.5: 41.2 on the Agentic Index · SpaceXAIGrok 4.541.2 on the Agentic Index · SpaceXAIDeepSeek V4 Flash 0731: 41.0 on the Agentic Index · DeepSeekDeepSeek V4 Flash 073141.0 on the Agentic Index · DeepSeekGemini 3.8 Flash: 40.2 on the Agentic Index · GoogleGemini 3.8 Flash40.2 on the Agentic Index · GoogleClaude Opus 4.7: 38.6 on the Agentic Index · AnthropicClaude Opus 4.738.6 on the Agentic Index · AnthropicGLM 5.2: 38.4 on the Agentic Index · Z.aiGLM 5.238.4 on the Agentic Index · Z.aiGPT-5.5: 36.4 on the Agentic Index · OpenAIGPT-5.536.4 on the Agentic Index · OpenAIGemini 3.7 Flash: 35.2 on the Agentic Index · GoogleGemini 3.7 Flash35.2 on the Agentic Index · Google

Every column stands on zero on the index's own 0 to 100 scale, so a column half as tall is half the score, and the maker's mark hangs under each; the leaderboard below carries every ranked model. Scores are Artificial Analysiss, as OpenRouter carries them.

9 models sit on the frontier: nothing scores higher for less.

  • the attractive quadrant
  • the frontier
  • a priced model, in its maker's tone
  • on the frontier, named
The attractive quadrant$0.10$1.00$10.000255075100Blended list price per M tokens, log scaleIndexClaude Fable 5.1: 57.2 · $20.00 blended · Anthropic$20.0057.209 Claude Fable 5.157.2 · $20.00 blended · AnthropicClaude Opus 5: 56.5 · $10.00 blended · Anthropic$10.0056.508 Claude Opus 556.5 · $10.00 blended · AnthropicQwen3.8 Max (0902): 56.0 · $3.00 blended · Qwen$3.0056.007 Qwen3.8 Max (0902)56.0 · $3.00 blended · QwenMuse Spark 1.3: 55.5 · $2.00 blended · Meta$2.0055.506 Muse Spark 1.355.5 · $2.00 blended · MetaGLM 5.3: 53.1 · $0.402 blended · Z.ai$0.40253.105 GLM 5.353.1 · $0.402 blended · Z.aiGrok 4.6: 52.5 · $3.00 blended · SpaceXAI$3.0052.5Grok 4.652.5 · $3.00 blended · SpaceXAIGPT-6 Astra: 51.0 · $20.00 blended · OpenAI$20.0051.0GPT-6 Astra51.0 · $20.00 blended · OpenAIGLM 5.3 Flash: 50.9 · $0.238 blended · Z.ai$0.23850.904 GLM 5.3 Flash50.9 · $0.238 blended · Z.aiClaude Fable 5: 50.7 · $20.00 blended · Anthropic$20.0050.7Claude Fable 550.7 · $20.00 blended · AnthropicQwen3.8 2.4T A95B: 50.1 · $3.00 blended · Qwen$3.0050.1Qwen3.8 2.4T A95B50.1 · $3.00 blended · QwenDeepSeek V4 Flash Vision Exp: 47.5 · $0.66 blended · DeepSeek$0.6647.5DeepSeek V4 Flash Vision Exp47.5 · $0.66 blended · DeepSeekGPT-5.6 Sol: 47.4 · $8.00 blended · OpenAI$8.0047.4GPT-5.6 Sol47.4 · $8.00 blended · OpenAIQwen3.8 27B: 45.8 · $0.581 blended · Qwen$0.58145.8Qwen3.8 27B45.8 · $0.581 blended · QwenClaude Sonnet 5: 43.6 · $4.00 blended · Anthropic$4.0043.6Claude Sonnet 543.6 · $4.00 blended · AnthropicMuse Spark 1.2: 43.2 · $2.00 blended · Meta$2.0043.2Muse Spark 1.243.2 · $2.00 blended · MetaGPT-5.6 Terra: 43.2 · $4.50 blended · OpenAI$4.5043.2GPT-5.6 Terra43.2 · $4.50 blended · OpenAIGPT-5.6 Luna: 42.1 · $0.45 blended · OpenAI$0.4542.1GPT-5.6 Luna42.1 · $0.45 blended · OpenAIClaude Opus 4.8: 41.9 · $10.00 blended · Anthropic$10.0041.9Claude Opus 4.841.9 · $10.00 blended · AnthropicDeepSeek V4 Pro 0813: 41.3 · $1.193 blended · DeepSeek$1.19341.3DeepSeek V4 Pro 081341.3 · $1.193 blended · DeepSeekGrok 4.5: 41.2 · $3.00 blended · SpaceXAI$3.0041.2Grok 4.541.2 · $3.00 blended · SpaceXAIDeepSeek V4 Flash 0731: 41.0 · $0.09 blended · DeepSeek$0.0941.003 DeepSeek V4 Flash 073141.0 · $0.09 blended · DeepSeekGemini 3.8 Flash: 40.2 · $1.65 blended · Google$1.6540.2Gemini 3.8 Flash40.2 · $1.65 blended · GoogleClaude Opus 4.7: 38.6 · $10.00 blended · Anthropic$10.0038.6Claude Opus 4.738.6 · $10.00 blended · AnthropicGLM 5.2: 38.4 · $1.705 blended · Z.ai$1.70538.4GLM 5.238.4 · $1.705 blended · Z.aiGPT-5.5: 36.4 · $11.25 blended · OpenAI$11.2536.4GPT-5.536.4 · $11.25 blended · OpenAIGemini 3.7 Flash: 35.2 · $1.50 blended · Google$1.5035.2Gemini 3.7 Flash35.2 · $1.50 blended · GoogleClaude Sonnet 4.6: 31.8 · $6.00 blended · Anthropic$6.0031.8Claude Sonnet 4.631.8 · $6.00 blended · AnthropicMiniMax M3: 29.5 · $0.525 blended · MiniMax$0.52529.5MiniMax M329.5 · $0.525 blended · MiniMaxKimi K3: 29.3 · $4.144 blended · MoonshotAI$4.14429.3Kimi K329.3 · $4.144 blended · MoonshotAIGemini 3.6 Flash: 29.0 · $1.50 blended · Google$1.5029.0Gemini 3.6 Flash29.0 · $1.50 blended · GoogleLing 3.0 Flash VL: 28.7 · $0.09 blended · inclusionAI$0.0928.7Ling 3.0 Flash VL28.7 · $0.09 blended · inclusionAIMotif 3: 28.4 · $0.525 blended · Motif Technologies$0.52528.4Motif 328.4 · $0.525 blended · Motif TechnologiesGemini 3.5 Flash: 26.0 · $3.375 blended · Google$3.37526.0Gemini 3.5 Flash26.0 · $3.375 blended · GoogleMuse Spark 1.1: 25.8 · $2.00 blended · Meta$2.0025.8Muse Spark 1.125.8 · $2.00 blended · MetaHy3 preview: 24.1 · $0.285 blended · Tencent$0.28524.1Hy3 preview24.1 · $0.285 blended · TencentGLM 5.1: 23.9 · $2.15 blended · Z.ai$2.1523.9GLM 5.123.9 · $2.15 blended · Z.aiInkling Small: 23.5 · $0.638 blended · Thinking Machines$0.63823.5Inkling Small23.5 · $0.638 blended · Thinking MachinesQwen3.7 Max: 22.5 · $2.213 blended · Qwen$2.21322.5Qwen3.7 Max22.5 · $2.213 blended · QwenInkling: 22.5 · $1.763 blended · Thinking Machines$1.76322.5Inkling22.5 · $1.763 blended · Thinking MachinesDeepSeek V4 Flash 0423: 22.2 · $0.113 blended · DeepSeek$0.11322.2DeepSeek V4 Flash 042322.2 · $0.113 blended · DeepSeekMiMo-V2.5-Pro: 21.3 · $0.544 blended · Xiaomi$0.54421.3MiMo-V2.5-Pro21.3 · $0.544 blended · XiaomiKimi K2.7 Code: 21.0 · $1.463 blended · MoonshotAI$1.46321.0Kimi K2.7 Code21.0 · $1.463 blended · MoonshotAIKimi K2.6: 20.5 · $1.34 blended · MoonshotAI$1.3420.5Kimi K2.620.5 · $1.34 blended · MoonshotAINemotron 3 Ultra: 20.1 · $0.925 blended · NVIDIA$0.92520.1Nemotron 3 Ultra20.1 · $0.925 blended · NVIDIALing 3.0 Flash: 19.3 · $0.032 blended · inclusionAI$0.03219.302 Ling 3.0 Flash19.3 · $0.032 blended · inclusionAIQwen3.6 27B: 18.5 · $1.04 blended · Qwen$1.0418.5Qwen3.6 27B18.5 · $1.04 blended · QwenGPT-5.4 Mini: 17.9 · $1.688 blended · OpenAI$1.68817.9GPT-5.4 Mini17.9 · $1.688 blended · OpenAIQwen3.7 Plus: 17.5 · $0.56 blended · Qwen$0.5617.5Qwen3.7 Plus17.5 · $0.56 blended · QwenGPT-5.4 Nano: 16.0 · $0.463 blended · OpenAI$0.46316.0GPT-5.4 Nano16.0 · $0.463 blended · OpenAIClaude Sonnet 4.5: 15.8 · $6.00 blended · Anthropic$6.0015.8Claude Sonnet 4.515.8 · $6.00 blended · AnthropicMiMo-V2.5: 15.8 · $0.80 blended · Xiaomi$0.8015.8MiMo-V2.515.8 · $0.80 blended · XiaomiGrok 4.3: 15.5 · $1.563 blended · SpaceXAI$1.56315.5Grok 4.315.5 · $1.563 blended · SpaceXAIMiniMax M2.7: 15.3 · $0.368 blended · MiniMax$0.36815.3MiniMax M2.715.3 · $0.368 blended · MiniMaxGemini 3.5 Flash Lite: 14.3 · $0.85 blended · Google$0.8514.3Gemini 3.5 Flash Lite14.3 · $0.85 blended · GoogleLongCat 2.0: 14.0 · $0.525 blended · Meituan$0.52514.0LongCat 2.014.0 · $0.525 blended · MeituanQwen3.6 35B A3B: 13.1 · $0.30 blended · Qwen$0.3013.1Qwen3.6 35B A3B13.1 · $0.30 blended · QwenMuse Glimmer 30B: 8.5 · $0.638 blended · Meta$0.6388.5Muse Glimmer 30B8.5 · $0.638 blended · MetaQwen3.5 397B A17B: 8.3 · $1.088 blended · Qwen$1.0888.3Qwen3.5 397B A17B8.3 · $1.088 blended · QwenGemini 3.1 Pro Preview: 8.2 · $4.50 blended · Google$4.508.2Gemini 3.1 Pro Preview8.2 · $4.50 blended · GoogleClaude Haiku 4.5: 8.0 · $2.00 blended · Anthropic$2.008.0Claude Haiku 4.58.0 · $2.00 blended · AnthropicQwen3.5-122B-A10B: 7.6 · $0.715 blended · Qwen$0.7157.6Qwen3.5-122B-A10B7.6 · $0.715 blended · QwenMistral Medium 3.5: 7.0 · $3.00 blended · Mistral$3.007.0Mistral Medium 3.57.0 · $3.00 blended · MistralDeepSeek V3.1 Terminus: 6.8 · $0.453 blended · DeepSeek$0.4536.8DeepSeek V3.1 Terminus6.8 · $0.453 blended · DeepSeekGPT-5 Mini: 6.8 · $0.688 blended · OpenAI$0.6886.8GPT-5 Mini6.8 · $0.688 blended · OpenAILing 3.0 Tiny: 4.4 · $0.023 blended · inclusionAI$0.0234.401 Ling 3.0 Tiny4.4 · $0.023 blended · inclusionAIGemma 4 31B: 4.2 · $0.25 blended · Google$0.254.2Gemma 4 31B4.2 · $0.25 blended · Googlegpt-oss-120b: 3.7 · $0.262 blended · OpenAI$0.2623.7gpt-oss-120b3.7 · $0.262 blended · OpenAINemotron 3.5 Lightning: 3.5 · $0.103 blended · NVIDIA$0.1033.5Nemotron 3.5 Lightning3.5 · $0.103 blended · NVIDIAGemma 4 26B A4B: 2.9 · $0.143 blended · Google$0.1432.9Gemma 4 26B A4B2.9 · $0.143 blended · GoogleDevstral 2 2512: 2.3 · $0.80 blended · Mistral$0.802.3Devstral 2 25122.3 · $0.80 blended · MistralMistral Medium 3.1: 1.9 · $0.88 blended · Mistral$0.881.9Mistral Medium 3.11.9 · $0.88 blended · MistralNemotron 3 Super: 1.7 · $0.156 blended · NVIDIA$0.1561.7Nemotron 3 Super1.7 · $0.156 blended · NVIDIAGemini 2.5 Pro: 1.6 · $3.438 blended · Google$3.4381.6Gemini 2.5 Pro1.6 · $3.438 blended · GoogleGemini 3.1 Flash Lite Preview: 1.6 · $0.563 blended · Google$0.5631.6Gemini 3.1 Flash Lite Preview1.6 · $0.563 blended · GoogleGranite 4.2 8B: 1.3 · $0.082 blended · IBM Granite$0.0821.3Granite 4.2 8B1.3 · $0.082 blended · IBM GraniteQwen3 235B A22B Thinking 2507: 1.3 · $1.213 blended · Qwen$1.2131.3Qwen3 235B A22B Thinking 25071.3 · $1.213 blended · Qwengpt-oss-20b: 1.2 · $0.036 blended · OpenAI$0.0361.2gpt-oss-20b1.2 · $0.036 blended · OpenAIQwen3.5-9B: 1.2 · $0.19 blended · Qwen$0.191.2Qwen3.5-9B1.2 · $0.19 blended · QwenMinistral 3 14B 2512: 1.1 · $0.20 blended · Mistral$0.201.1Ministral 3 14B 25121.1 · $0.20 blended · MistralCommand A+: 1.0 · $0.60 blended · Cohere$0.601.0Command A+1.0 · $0.60 blended · CohereMistral Large 3 2512: 1.0 · $0.75 blended · Mistral$0.751.0Mistral Large 3 25121.0 · $0.75 blended · MistralNemotron 3 Nano 30B A3B: 1.0 · $0.105 blended · NVIDIA$0.1051.0Nemotron 3 Nano 30B A3B1.0 · $0.105 blended · NVIDIAQwen3 14B: 0.9 · $0.15 blended · Qwen$0.150.9Qwen3 14B0.9 · $0.15 blended · QwenQwen3 32B: 0.9 · $0.13 blended · Qwen$0.130.9Qwen3 32B0.9 · $0.13 blended · QwenQwen3 Coder Next: 0.9 · $0.29 blended · Qwen$0.290.9Qwen3 Coder Next0.9 · $0.29 blended · QwenDeepSeek V3 0324: 0.8 · $0.502 blended · DeepSeek$0.5020.8DeepSeek V3 03240.8 · $0.502 blended · DeepSeekMinistral 3 3B 2512: 0.8 · $0.10 blended · Mistral$0.100.8Ministral 3 3B 25120.8 · $0.10 blended · MistralMistral Small 4: 0.8 · $0.262 blended · Mistral$0.2620.8Mistral Small 40.8 · $0.262 blended · MistralLlama 4 Maverick: 0.6 · $0.513 blended · Meta Llama$0.5130.6Llama 4 Maverick0.6 · $0.513 blended · Meta LlamaMinistral 3 8B 2512: 0.6 · $0.15 blended · Mistral$0.150.6Ministral 3 8B 25120.6 · $0.15 blended · MistralLlama 4 Scout: 0.5 · $0.15 blended · Meta Llama$0.150.5Llama 4 Scout0.5 · $0.15 blended · Meta LlamaGemma 3 12B: 0.1 · $0.075 blended · Google$0.0750.1Gemma 3 12B0.1 · $0.075 blended · GoogleGemma 3 27B: 0.1 · $0.148 blended · Google$0.1480.1Gemma 3 27B0.1 · $0.148 blended · Google
  1. 01Ling 3.0 Tiny4.4 · $0.023 blended
  2. 02Ling 3.0 Flash19.3 · $0.032 blended · +14.9 at 1.4x
  3. 03DeepSeek V4 Flash 073141.0 · $0.09 blended · +21.7 at 2.9x
  4. 04GLM 5.3 Flash50.9 · $0.238 blended · +9.9 at 2.6x
  5. 05GLM 5.353.1 · $0.402 blended · +2.2 at 1.7x
  6. 06Muse Spark 1.355.5 · $2.00 blended · +2.4 at 5.0x
  7. 07Qwen3.8 Max (0902)56.0 · $3.00 blended · +0.5 at 1.5x
  8. 08Claude Opus 556.5 · $10.00 blended · +0.5 at 3.3x
  9. 09Claude Fable 5.157.2 · $20.00 blended · +0.7 at 2.0x

93 of 94 ranked models carry a list price; 1 is free or unlisted and not in the field. Price is the blended OpenRouter rate, three input tokens to one output, on a logarithmic axis. Each dot is filled in its maker's tone, as the columns above are. The numbered and named dots are the frontier, cheapest first; each step after the first names what it adds over the step before, and at what multiple of the blended price. A hover over any dot names it and its two figures, a click opens the model, and a model named in the picker above keeps its label on. 14 of 93 priced models sit in the attractive quadrant: at or above the median score of 17.9 and at or below the median blended price of $0.715. The tinted corner is that quadrant. 8 of the 93 crosses match the selection below and are drawn white.

8 of 94.

Artificial Analysis · 0 to 100
PositionModelAuthorScoreIn / out per MContextListed
62Mistral Medium 3.5Mistral7.0$1.50 / $7.50262k
70Devstral 2 2512Mistral2.3$0.40 / $2.00262k
71Mistral Medium 3.1Mistral1.9$0.44 / $2.20131k
80Ministral 3 14B 2512Mistral1.1$0.20 / $0.20262k
82Mistral Large 3 2512Mistral1.0$0.50 / $1.50262k
88Ministral 3 3B 2512Mistral0.8$0.10 / $0.10131k
89Mistral Small 4Mistral0.8$0.15 / $0.60262k
91Ministral 3 8B 2512Mistral0.6$0.15 / $0.15262k

The field and the frontier above are the whole board and do not narrow with this table: a frontier drawn through 8 of 94 models would say nothing beats them on both while something in the same board does. The models this table holds are drawn white in the figures instead.

What the score tells you

A composite the publisher computes from its own evaluation runs. It orders models on the publisher's tasks, not yours; a gap of a point or two is within what a different task mix would move. It says nothing about price, latency, or how a model behaves on your own cases, which is where Vitna's engine measures quality.

Sources

Source: OpenRouter (openrouter.ai/models), as of 17 Nov 16:29 UTC. sha256 0bb53a77713f over 853,270 bytes.
Catalog: OpenRouter's public model list.
Intelligence, coding and agentic indexes: Artificial Analysis, as published by OpenRouter.
Elo, win rate and rank: Design Arena, as published by OpenRouter.
Vitna ran none of these evaluations and adjusted none of these figures.