Category scores
Out of 100: 100 = the best tracked model in the category.
| Reasoning | 99 | |
| Mathematics | 96 | |
| Code | 97 | |
| Agents & real work | 86 | |
| Knowledge | 100 | |
| Long context | 95 | |
| Multimodal | 100 |
EditionUpdated daily
Ranking / Claude Opus 5.5
Claude Opus 5.5
Anthropic · Proprietary · context 1M. Ranked #2 of 26 with 921 ERAZ points out of 1000.
ERAZ rank#2/ 26
Points921/ 1000
Coverage85%
Benchmarks won4/ 14
Input price / M$4
Output price / M$20
Points over time
| Day | Points | Change |
|---|
Results on each benchmark
| Benchmark | Score | Rank | Out of 100 | Source |
|---|---|---|---|---|
| Humanity's Last Exam | 61.4% | #1 / 26 | 100 | src ↗ |
| ARC-AGI-2 | 91.7% | #4 / 13 | 97 | src ↗ |
| ARC-AGI-3 | not published | — | — | |
| FrontierMath (Tiers 1-3) | 91.2% | #3 / 16 | 97 | src ↗ |
| FrontierMath (Tier 4) | 95% | #3 / 16 | 95 | src ↗ |
| SWE-bench Pro | 89.9% | #1 / 9 | 100 | src ↗ |
| FrontierSWE v2 | 62.3% | #2 / 14 | 95 | src ↗ |
| Terminal-Bench 2.1 | not published | — | — | |
| OSWorld 2.0 | 48.7% | #8 / 11 | 67 | src ↗ |
| GDPval-AA | 1846 | #3 / 23 | 98 | src ↗ |
| GPQA Diamond | not published | — | — | |
| AA-Omniscience | 46.4 | #1 / 26 | 100 | src ↗ |
| AA-LCR | 84.7% | #5 / 26 | 95 | src ↗ |
| MMMU-Pro | 87.7% | #1 / 14 | 100 | src ↗ |
Artificial Analysis Intelligence Index : 57.6 · Source page ↗
Nearby models in the ranking
</> Embed this ranking
Show Claude Opus 5.5's ranking on your site or in a README. The badge updates every day.
