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 June 22, 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 Fable 5 | 59.9 | #50 100% | 2.00 |
| 2 | Claude Opus 4.8 | 55.7 | #9 17% | 3.71 |
| 3 | GPT-5.5 | 54.8 | #16 101% | 3.13 |
| 4 | Claude Opus 4.7 | 53.5 | #5 18% | 3.57 |
| 5 | GPT-5.4 | 51.4 | #28 10% | 5.87 |
| 6 | GLM-5.2 | 51.1 | #27 — | 17.62 |
| 7 | Gemini 3.5 Flash | 50.2 | #21 26% | 9.56 |
| 8 | Claude Sonnet 4.6 | 47.2 | #7 36% | 5.24 |
| 9 | Gemini 3.1 Pro Preview | 46.5 | #30 32% | 6.64 |
| 10 | Qwen3.7 Max | 46.0 | #39 44% | 9.20 |
| 11 | MiniMax-M3 | 44.4 | #4 32% | 59.20 |
| 12 | GPT-5.3 Codex | 44.3 | #44 59% | 5.63 |
| 13 | DeepSeek V4 Pro | 44.3 | #6 17% | 68.15 |
| 14 | Claude Opus 4.6 | 43.7 | #23 20% | 2.91 |
| 15 | Muse Spark | 43.1 | — | — |
| 16 | Kimi K2.6 | 42.8 | #20 13% | 17.29 |
| 17 | MiMo-V2.5-Pro | 42.2 | #11 14% | 64.92 |
| 18 | GPT-5.2 | 42.2 | #62 — | 5.36 |
| 19 | Kimi K2.7 Code | 41.9 | #48 85% | 16.93 |
| 20 | Claude Opus 4.5 | 40.8 | #69 — | 2.72 |
| 21 | MiMo-V2-Pro | 40.3 | — | 20.15 |
| 22 | DeepSeek V4 Flash | 40.3 | #1 0% | 191.90 |
| 23 | GLM-5.1 | 40.2 | #18 73% | 13.86 |
| 24 | GPT-5.2 Codex | 40.1 | — | 5.09 |
| 25 | MiMo-V2.5 | 40.1 | #3 7% | 190.95 |
| 26 | GPT-5.4 mini | 40.0 | #35 8% | 15.24 |
| 27 | Qwen3.6 Max Preview | 40.0 | — | 8.79 |
| 28 | Gemini 3 Pro Preview | 39.6 | — | 5.66 |
| 29 | Qwen3.6 Plus | 39.6 | — | 22.63 |
| 30 | GLM-5 | 39.5 | #40 14% | 18.81 |
| 31 | Qwen3.7 Plus | 39.0 | #51 30% | 50.00 |
| 32 | GPT-5.1 | 38.9 | #56 100% | 6.92 |
| 33 | GPT-5.4 nano | 38.2 | #46 24% | 52.69 |
| 34 | GLM-5-Turbo | 38.1 | — | — |
| 35 | MiniMax-M2.7 | 38.1 | #26 3% | 50.80 |
| 36 | Kimi K2.5 | 38.1 | — | 21.28 |
| 37 | Nemotron 3 Ultra 550B A55B | 37.8 | #24 34% | 22.57 |
| 38 | Gemini 3 Flash Preview | 37.8 | #10 17% | 21.60 |
| 39 | Grok 4.3 | 37.6 | #59 100% | 20.05 |
| 40 | Qwen3.6 27B | 37.1 | — | 17.67 |
| 41 | Grok 4.20 0309 v2 | 37.0 | — | 9.25 |
| 42 | Grok 4.20 0309 | 36.5 | — | 9.13 |
| 43 | MiMo-V2-Omni-0327 | 36.4 | — | 30.33 |
| 44 | GPT-5 | 36.1 | #58 — | 6.42 |
| 45 | GPT-5 Codex | 36.1 | — | 6.42 |
| 46 | MiMo-V2-Omni | 35.0 | — | 3500.00 |
| 47 | GPT-5.1 Codex | 34.7 | — | 6.17 |
| 48 | Claude 4.5 Sonnet | 34.7 | #45 27% | 3.86 |
| 49 | GLM 5V Turbo | 34.5 | — | — |
| 50 | Qwen3.5 27B | 33.8 | — | 25.04 |
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.