Public KG record
Terminal-Bench 2.1
Recorded overview
Terminal command benchmark; Granite 4.2 30B scored 29.24.
An evidence-backed entity from the project knowledge graph; detailed tool metadata is still being added.
Recorded details
- Benchmarks: best result +4.9 quality points at 67% lower cost
- Benchmarks: DeepSeek-V4.1-Flash 在 Terminal-Bench 2.1 上超过 Opus-5 与 GPT-5.6 Sol
- Benchmarks: Creation使用89个实例
- Benchmarks: Evolution固定反馈集包含全部89个任务
- Benchmarks: 包含 89 个实例
- Benchmarks: Evolution 反馈集包含全部 89 个任务
- Benchmarks: 人类工程参考得分为 88.8
- Benchmarks: GPT-5 在 Terminus 2 内解决 35.2% 的任务
Evidence coverage
The KG currently records 5 known information dimensions and has 12 undocumented dimensions, linked to 23 public source articles. These counts describe record coverage, not model performance or verified fit.
Related public records
- Claude Code — evaluated on
- Strands Harness — evaluated on
- HarnessDev — evaluated on
- JOLT — evaluated on
- RRSI — evaluated on