AA Agentic Index
| # | Model | Score | Observed | |
|---|---|---|---|---|
| 1 | GPT-5.6 SolOpenAI | 54.0 mirror | Jul 23, 2026 | |
| 2 | Claude Fable 5Anthropic | 52.8 mirror | Jul 23, 2026 | |
| 3 | Kimi K3Moonshot AI | 50.1 mirror | Jul 23, 2026 | |
| 4 | GPT-5.6 TerraOpenAI | 47.4 mirror | Jul 23, 2026 | |
| 5 | Claude Opus 4.8Anthropic | 47.2 mirror | Jul 23, 2026 | |
| 6 | Claude Sonnet 5Anthropic | 46.7 independent | Jul 24, 2026 | |
| 7 | Grok 4.5xAI | 45.7 mirror | Jul 23, 2026 | |
| 8 | GPT-5.6 LunaOpenAI | 45.6 independent | Jul 24, 2026 | |
| 9 | GPT-5.5OpenAI | 44.9 independent | Jul 24, 2026 | |
| 10 | Claude Opus 4.7Anthropic | 44.4 mirror | Jul 23, 2026 | |
| 11 | GLM-5.2Zhipu AI | 43.1 mirror | Jul 23, 2026 | |
| 12 | GPT-5.4OpenAI | 41.1 mirror | Jul 23, 2026 | |
| 13 | Claude Sonnet 4.6Anthropic | 40.8 mirror | Jul 23, 2026 | |
| 14 | Gemini 3.6 FlashGoogle | 38.7 mirror | Jul 23, 2026 | |
| 15 | Muse Spark 1.1Meta | 37.5 mirror | Jul 23, 2026 | |
| 16 | Gemini 3.5 FlashGoogle | 37.4 independent | Jul 24, 2026 | |
| 17 | DeepSeek V4 ProDeepSeek | 36.4 mirror | Jul 23, 2026 | |
| 18 | MiniMax-M3MiniMax (minimax.io) | 35.4 independent | Jul 24, 2026 | |
| 19 | DeepSeek V4 FlashDeepSeek | 31.1 mirror | Jul 23, 2026 | |
| 20 | Qwen3.7 MaxAlibaba | 30.6 mirror | Jul 23, 2026 | |
| 21 | Kimi K2.6Moonshot AI | 30.3 mirror | Jul 23, 2026 | |
| 22 | GPT-5.4 miniOpenAI | 30.2 mirror | Jul 23, 2026 | |
| 23 | GLM-5.1Zhipu AI | 29.9 mirror | Jul 23, 2026 | |
| 24 | Kimi K2.7 CodeMoonshot AI | 29.6 mirror | Jul 23, 2026 | |
| 25 | Grok Build 0.1xAI | 28.0 independent | Jul 24, 2026 | |
| 26 | Qwen3.6 PlusAlibaba | 27.6 mirror | Jul 23, 2026 | |
| 27 | GPT-5.4 nanoOpenAI | 27.5 mirror | Jul 23, 2026 | |
| 28 | Nemotron 3 Ultra 550B A55BNvidia | 27.4 mirror | Jul 23, 2026 | |
| 29 | Qwen3.6 27BAlibaba | 27.0 mirror | Jul 23, 2026 | |
| 30 | Gemini 3.5 Flash LiteGoogle | 26.8 independent | Jul 24, 2026 | |
| 31 | GPT-5OpenAI | 25.7 independent | Jul 24, 2026 | |
| 32 | MiniMax-M2.7MiniMax (minimax.io) | 25.6 independent | Jul 24, 2026 | |
| 33 | GLM-4.7Zhipu AI | 25.4 mirror | Jul 23, 2026 | |
| 34 | Claude Sonnet 4.5 (latest)Anthropic | 24.6 mirror | Jul 23, 2026 | |
| 35 | Grok 4.3xAI | 24.1 mirror | Jul 23, 2026 | |
| 36 | Kimi K2.5Moonshot AI | 21.7 mirror | Jul 23, 2026 | |
| 37 | Step 3.7 FlashStepFun (China) | 21.5 mirror | Jul 23, 2026 | |
| 38 | Gemini 3.1 Pro PreviewGoogle | 21.4 mirror | Jul 23, 2026 | |
| 38 | Qwen3.6 35B-A3BAlibaba | 21.4 mirror | Jul 23, 2026 | |
| 40 | GPT-5.1OpenAI | 21.0 independent | Jul 24, 2026 | |
| 41 | Qwen3.7 PlusAlibaba | 20.8 mirror | Jul 23, 2026 | |
| 42 | Qwen3.5 122B-A10BAlibaba | 20.7 independent | Jul 24, 2026 | |
| 43 | Qwen3.5 397B-A17BAlibaba | 19.8 independent | Jul 24, 2026 | |
| 44 | GPT-5 MiniOpenAI | 19.4 independent | Jul 24, 2026 | |
| 45 | DeepSeek V3.2DeepSeek | 18.3 mirror | Jul 23, 2026 | |
| 46 | GLM-4.6Zhipu AI | 17.7 mirror | Jul 23, 2026 | |
| 47 | Claude Haiku 4.5 (latest)Anthropic | 16.4 mirror | Jul 23, 2026 | |
| 48 | Gemma 4 31B ITGoogle | 14.4 mirror | Jul 23, 2026 | |
| 49 | Qwen3.5 35B-A3BAlibaba | 11.8 mirror | Jul 23, 2026 | |
| 50 | Gemma 4 26B A4B ITGoogle | 11.0 mirror | Jul 23, 2026 | |
| 51 | Devstral 2Mistral | 10.6 mirror | Jul 23, 2026 | |
| 52 | Mercury 2Inception | 9.6 independent | Jul 24, 2026 | |
| 53 | Nemotron 3 SuperNvidia | 8.7 mirror | Jul 23, 2026 | |
| 54 | Gemini 2.5 ProGoogle | 7.1 independent | Jul 24, 2026 | |
| 55 | Gemini 3.1 Flash Lite PreviewGoogle | 6.2 mirror | Jul 23, 2026 | |
| 56 | Qwen3 235B-A22BthinkingAlibaba | 3.8 mirror | Jul 23, 2026 | |
| 57 | DeepSeek R1DeepSeek | 3.1 independent | Jul 24, 2026 | |
| 58 | Qwen3-Next 80B-A3B (Thinking)thinkingAlibaba | 2.1 mirror | Jul 23, 2026 | |
| 59 | nemotron-3-nano-30b-a3bNvidia | 2.0 mirror | Jul 23, 2026 | |
| 60 | Qwen3 32BAlibaba | 1.8 mirror | Jul 23, 2026 | |
| 60 | Qwen3 14BAlibaba | 1.8 mirror | Jul 23, 2026 | |
| 60 | Kimi K2 ThinkingthinkingMoonshot AI | 1.8 mirror | Jul 23, 2026 | |
| 63 | GPT-4.1 miniOpenAI | 1.7 mirror | Jul 23, 2026 | |
| 63 | o3-minihigh-effortOpenAI | 1.7 independent | Jul 24, 2026 | |
| 65 | Qwen3 8BAlibaba | 1.5 mirror | Jul 23, 2026 | |
| 66 | GPT-4.1 nanoOpenAI | 1.2 mirror | Jul 23, 2026 | |
| 67 | GPT-4o miniOpenAI | 1.0 mirror | Jul 23, 2026 | |
| 68 | Llama-3.3-70B-InstructMeta | 0.3 mirror | Jul 23, 2026 |
Bars on a 0–54 index scale. Scores link to model pages; every number keeps its source on the model page.