AI Model Leaderboard
The only board that ranks LLMs on all three axes that matter to a builder: how smart they are (intelligence), how widely used they are (real OpenRouter traffic), and your value — intelligence per dollar.
Updated August 10, 2026
Ranked by the Artificial Analysis Intelligence Index. Value = intelligence per blended $/1M token. Popularity shown where the model has OpenRouter traffic.
| # | Model | |||
|---|---|---|---|---|
| 1 | Claude Opus 5 | 63.1 | #19 8% | 4.21 |
| 2 | Claude Fable 5 | 62.1 | #33 13% | 2.07 |
| 3 | GPT-5.6 Sol | 60.9 | #24 18% | 3.48 |
| 4 | Kimi K3 | 59.7 | #11 14% | 6.63 |
| 5 | Qwen3.8 Max | 58.1 | #58 — | 14.53 |
| 6 | Claude Opus 4.8 | 57.3 | #10 14% | 3.82 |
| 7 | Muse Spark 1.2 | 56.8 | — | 20.65 |
| 8 | GPT-5.6 Terra | 56.6 | #27 22% | 8.09 |
| 9 | GPT-5.5 | 56.3 | #31 30% | 3.22 |
| 10 | Grok 4.5 | 55.8 | #26 19% | 13.95 |
| 11 | Claude Sonnet 5 | 55.3 | #12 14% | 9.22 |
| 12 | Claude Opus 4.7 | 55.0 | #14 57% | 3.67 |
| 13 | Muse Spark 1.1 | 53.2 | — | 19.35 |
| 14 | GPT-5.4 | 53.1 | #35 53% | 6.07 |
| 15 | GLM-5.2 | 52.6 | #4 7% | 18.65 |
| 16 | GPT-5.6 Luna | 52.3 | #8 58% | 74.71 |
| 17 | Gemini 3.5 Flash | 52.0 | #34 37% | 9.90 |
| 18 | DeepSeek V4 Flash 0731 | 51.8 | — | 246.67 |
| 19 | Gemini 3.6 Flash | 51.6 | #17 421% | 11.47 |
| 20 | Claude Sonnet 4.6 | 48.4 | #15 30% | 5.38 |
| 21 | Gemini 3.1 Pro Preview | 47.7 | #44 37% | 6.81 |
| 22 | Qwen3.7 Max | 46.7 | #56 — | 9.34 |
| 23 | GPT-5.3 Codex | 45.5 | — | 5.78 |
| 24 | MiniMax-M3 | 45.4 | #7 24% | 60.53 |
| 25 | DeepSeek V4 Pro | 45.3 | #5 30% | 69.69 |
| 26 | Kimi K2.6 | 45.1 | #38 100% | 18.22 |
| 27 | Claude Opus 4.6 | 44.9 | #43 38% | 2.99 |
| 28 | Motif 3 | 44.9 | — | — |
| 29 | Muse Spark | 43.5 | — | — |
| 30 | GPT-5.2 | 43.3 | #61 — | 5.50 |
| 31 | Kimi K2.7 Code | 43.0 | #57 — | 17.37 |
| 32 | MiMo-V2.5-Pro | 42.9 | #21 27% | 66.00 |
| 33 | Inkling | 42.3 | — | 16.75 |
| 34 | Hy3 | 42.2 | #3 47% | 120.57 |
| 35 | DeepSeek V4 Flash | 42.1 | #1 35% | 205.37 |
| 36 | Claude Opus 4.5 | 41.9 | — | 2.79 |
| 37 | Nex-N2-Pro | 41.7 | — | 27.80 |
| 38 | MiMo-V2-Pro | 41.4 | — | — |
| 39 | GPT-5.2 Codex | 41.2 | — | 5.23 |
| 40 | Inkling Small | 41.2 | — | 54.93 |
| 41 | Qwen3.6 Max Preview | 41.1 | — | 9.03 |
| 42 | GLM-5.1 | 41.0 | #65 — | 14.19 |
| 43 | GPT-5.4 mini | 40.9 | #47 38% | 15.58 |
| 44 | Grok Build 0.1 0616 | 40.7 | — | 27.13 |
| 45 | GLM-5 | 40.6 | #45 65% | 19.33 |
| 46 | Gemini 3 Pro Preview | 40.6 | — | 5.80 |
| 47 | Qwen3.6 Plus | 40.5 | — | 23.14 |
| 48 | JT-4.1 Flash 236B A21B | 39.8 | — | — |
| 49 | GPT-5.4 nano | 39.7 | #46 100% | 54.76 |
| 50 | Agnes 2.5 Pro Alpha | 39.7 | — | 58.81 |
The board is a superset of the calculator catalog: brand-new models appear here with quality & popularity before the pricing catalog catches up. Models with a calculator match link to a head-to-head comparison.
Data: Artificial Analysis · OpenRouter
Know your margin before you build
Pick a model above, then model your real cost and price in the Calcaas calculator — free.