Public KG record
Claude Opus 5.5
Recorded overview
Achieves 67.3% pass@1 and 78.0% pass@4 on CWE-bench v2 at $1.83 per rollout.
An evidence-backed entity from the project knowledge graph; detailed tool metadata is still being added.
Recorded details
- Benchmarks: CWE-bench v1得分67%
- Benchmarks: Artificial Analysis Intelligence Index v4.3：58分
- Benchmarks: Terminal-Bench 4.0：66.4%
- Benchmarks: SWE-Marathon：46.3
- Benchmarks: Terminal-Bench Science 0.1：63.3
- Benchmarks: BenchCAD：有工具96.2，无工具73.0
- Benchmarks: Agents’ Last Exam v1：63.2%，$8.82／任务
- Benchmarks: Terminal-Bench-Science 0.1：63.3%，$23.21／任务
Evidence coverage
The KG currently records 14 known information dimensions and has 5 undocumented dimensions, linked to 39 public source articles. These counts describe record coverage, not model performance or verified fit.
Related public records
- Claude Code — uses
- Droid 1.0 — supports
- Exa Agent Ultra — outperforms
- GDPx benchmark — evaluated on
- StateBench — evaluated on