Public KG record
Opus 5
Recorded overview
An AI model, used in conjunction with Prime Agent to achieve high performance on
An evidence-backed entity from the project knowledge graph; detailed tool metadata is still being added.
Recorded details
- Benchmarks: 在 Terminal-Bench 2.1 与 DeepSWE v1.1 上被 DeepSeek-V4.1-Flash 超过
- Benchmarks: EdgeBench 上作为 SoL-Pi 后端
- Benchmarks: Terminal Bench 4.0得分51.8
- Benchmarks: Model ML Composite
- Benchmarks: PI评估
- Capabilities: 提示注入
- Performance: 52.6% on Terminal-Bench-Science 0.1 (comparison: 29.0%)
- Performance: 代码迁移使用的token数量为Claude Opus 5.5的2.5倍
Evidence coverage
The KG currently records 8 known information dimensions and has 10 undocumented dimensions, linked to 11 public source articles. These counts describe record coverage, not model performance or verified fit.
Related public records
- Anthropic — releases
- Ramp AI index — uses
- Claude Fable 5.1 — outperforms
- Claude Opus 5.5 — outperforms
- GPT-5.6 Sol — competes with