Category scores
Out of 100: 100 = the best tracked model in the category.
| Reasoning | 77 | |
| Mathematics | 80 | |
| Code | 87 | |
| Agents & real work | 99 | |
| Knowledge | 85 | |
| Long context | 89 | |
| Multimodal | 96 |
EditionUpdated daily
Ranking / Claude Opus 5
Claude Opus 5
Anthropic · Proprietary · context not disclosed. Ranked #3 of 26 with 870 ERAZ points out of 1000.
ERAZ rank#3/ 26
Points870/ 1000
Coverage100%
Benchmarks won1/ 14
Input price / M—
Output price / M—
Points over time
| Day | Points | Change |
|---|
Results on each benchmark
| Benchmark | Score | Rank | Out of 100 | Source |
|---|---|---|---|---|
| Humanity's Last Exam | 54.9% | #5 / 26 | 89 | src ↗ |
| ARC-AGI-2 | 90.4% | #6 / 13 | 95 | src ↗ |
| ARC-AGI-3 | 30.2% | #4 / 9 | 48 | src ↗ |
| FrontierMath (Tiers 1-3) | 85.6% | #8 / 16 | 91 | src ↗ |
| FrontierMath (Tier 4) | 73.2% | #8 / 16 | 73 | src ↗ |
| SWE-bench Pro | 79.2% | #4 / 9 | 88 | src ↗ |
| FrontierSWE v2 | 52% | #6 / 14 | 79 | src ↗ |
| Terminal-Bench 2.1 | 84.6% | #4 / 12 | 97 | src ↗ |
| OSWorld 2.0 | 70.6% | #2 / 11 | 97 | src ↗ |
| GDPval-AA | 1862 | #1 / 23 | 100 | src ↗ |
| GPQA Diamond | 93.2% | #9 / 15 | 95 | src ↗ |
| AA-Omniscience | 37.1 | #6 / 26 | 80 | src ↗ |
| AA-LCR | 79.3% | #25 / 26 | 89 | src ↗ |
| MMMU-Pro | 84.7% | #5 / 14 | 96 | src ↗ |
Artificial Analysis Intelligence Index : — · Source page ↗
Nearby models in the ranking
</> Embed this ranking
Show Claude Opus 5's ranking on your site or in a README. The badge updates every day.
