Claude Opus 4.5
#53 of 274 scored.
AnthropicNov 24, 2025Closed
Capabilities Index
Range 148.0–152.8
150.1
Context
Takes file, image, text
200K
Input
Per 1M tokens
$5.00
Output
Per 1M tokens
$25
Benchmarks
How it scores.
- GPQA Diamond81.4%
60th of 186
- SWE-bench Verified76.6%
9th of 32
- FrontierMath34.4%
57th of 81
- ARC-AGI-237.6%
33rd of 79
- Humanity's Last Exam21.4%
14th of 40
- SimpleQA Verified45.7%
37th of 77
- Mock AIME86.1%
68th of 176
- Terminal-Bench63.1%
10th of 35
Everything else Epoch has for it
- DTBench83.1%
- Cybench82.0%
- ARC-AGI80.0%
- GeoBench75.0%
- METR Time Horizons75.0%
- OSWorld66.3%
- WeirdML63.7%
- DeepResearch Bench54.8%
- SimpleBench54.4%
- LMCA52.3%
- GDPval45.5%
- Balrog43.5%
- ProofBench36.0%
- GSO-Bench26.5%
- CL-bench21.1%
- EBR-bench14.3%
- Mystery Game Puzzles14.1%
- VPCT10.0%
- Chess Puzzles7.4%
- FrontierMath-Tier-4-v2-Private4.9%
- Remote Labor Index3.8%
- Furniture Assembly0.0%
On the index
Around it.
Sources