AA Intelligence Index v5
Intelligence index · 17 models · source Artificial Analysis · updated 2026-10-27
What this benchmark measures
A weighted composite of 10 independent evaluations covering knowledge, reasoning, coding, agents and long-horizon tasks. v5 went live on 27 October with a stricter scale than v4.3.2; scores from the two versions are not comparable.
Why it gets its own table
The only composite independent index that covers closed and open models and refreshes every two weeks. Its weakness is that the weights are not public, so this site uses it as the primary sort while showing the component scores alongside.
Full ranking
| # | Model | Score | Vs. best | Input $ | Output $ | Type |
|---|---|---|---|---|---|---|
| 1 | Gemini 4 CarbonGoogle | 59.4 | Best overall | $2.00 | $10.00 | |
| 2 | Claude Opus 5.5Anthropic | 58.4 | 98% | $4.00 | $20.00 | |
| 3 | GPT-6 AstraOpenAI | 56.9 | 96% | $10.00 | $50.00 | |
| 4 | GPT-6.1 SolOpenAI | 54.5 | 92% | $2.00 | $10.00 | |
| 5 | Claude Sonnet 5.5Anthropic | 54.2 | 91% | $2.00 | $10.00 | |
| 6 | GPT-6.1 TerraOpenAI | 51.0 | 86% | $1.00 | $5.00 | |
| 7 | Gemini 4 ArgonGoogle | 50.4 | 85% | $2.00 | $10.00 | |
| 8 | Claude Fable 5.1Anthropic | 49.7 | 84% | $10.00 | $50.00 | |
| 9 | Kimi K3.1 Pro Open weightsMoonshot AI | 46.0 | 77% | $2.00 | $10.00 | Open weights |
| 10 | Muse WonderMeta | 43.3 | 73% | $2.00 | $10.00 | |
| 11 | Claude Haiku 5.5Anthropic | 36.2 | 61% | $0.10 | $0.50 | |
| 12 | Grok 4.7xAI | 35.0 | 59% | $2.00 | $6.00 | |
| 13 | DeepSeek V4.1 Flash Open weightsDeepSeek | 32.5 | 55% | $0.30 | $1.20 | Open weights |
| 14 | GLM-5.3 Open weightsZ.ai | 32.0 | 54% | $0.60 | $2.20 | Open weights |
| 15 | MiMo-V2.6 Pro Open weightsXiaomi | 30.0 | 51% | $0.35 | $1.40 | Open weights |
| 16 | MiniMax M3.1 FlashMiniMax | 29.0 | 49% | $0.15 | $0.60 | |
| 17 | DeepSeek V4 Pro Open weightsDeepSeek | 23.0 | 39% | $1.32 | $3.96 | Open weights |