IQ vs Task Cost
Each model's estimated IQ plotted against the cost of a representative 1M-token task. Toggle between a coding task (cache-heavy — 800K cache-read + 100K input + 100K output) and a copywriting task (output-heavy — 150K input + 850K output).
IQ vs Task Cost
Each model's estimated IQ plotted against the cost of a representative 1M-token task. Toggle between a coding task (cache-heavy — 800K cache-read + 100K input + 100K output) and a copywriting task (output-heavy — 150K input + 850K output).
How To Read This Chart
This chart uses derived AI IQ fields and source-backed benchmark data from the public AI IQ dataset.
Top Models
| Rank | Model | Provider | Score |
|---|---|---|---|
| 1 | fable-5 | Anthropic | 132 |
| 2 | opus-4.8 | Anthropic | 130 |
| 3 | gpt-5.5 | OpenAI | 129 |
| 4 | opus-4.7 | Anthropic | 128 |
| 5 | gpt-5.4 | OpenAI | 124 |
| 6 | gemini-3.1-pro | 123 | |
| 7 | opus-4.6 | Anthropic | 123 |
| 8 | sonnet-5 | Anthropic | 123 |
| 9 | gemini-3.5-flash | 121 | |
| 10 | sonnet-4.6 | Anthropic | 119 |
| 11 | glm-5.2 | Z.ai | 117 |
| 12 | kimi-k2.6 | Kimi | 117 |