Anthropic
opus-4
Public AI IQ model profile with source-backed benchmark results and derived ranking context. Proprietary
IQ
n/a
EQ
n/a
Effective Cost
n/a
IQ Rank
n/a
IQ Dimensions
| Dimension | IQ | Coverage |
|---|---|---|
| mathematical reasoning | 96 | 4/5 |
| scientific reasoning | 113 | 3/4 |
| abstract reasoning | 77 | 2/3 |
| app building | n/a | 0/4 |
| production engineering | 94 | 2/7 |
| computer use | n/a | 0/7 |
| reliability | 89 | 1/3 |
Benchmark Results
| Benchmark | Score | Dimension |
|---|---|---|
| AIME | 41.25 | mathematical-reasoning |
| ARC-AGI-1 | 35.7 | abstract-reasoning |
| ARC-AGI-2 | 8.6 | abstract-reasoning |
| BullshitBench v2 | 34 | reliability |
| FrontierMath Tier 1-3 | 4.1 | mathematical-reasoning |
| FrontierMath Tier 4 | 4.2 | mathematical-reasoning |
| GPQA Diamond | 71.7 | scientific-reasoning |
| Humanity's Last Exam | 10.7 | scientific-reasoning |
| LiveCodeBench | 70.2 | production-engineering |
| MathArena | 31.6 | mathematical-reasoning |
| SciCode | 39.8 | scientific-reasoning |
| SWE-Bench Verified | 73.2 | production-engineering |