Public KG record
Gemini 4 Argon
Recorded overview
Leads CWE-bench v2 with 67.8% pass@1 and 81.0% pass@4 at $3.92 per rollout.
An evidence-backed entity from the project knowledge graph; detailed tool metadata is still being added.
Recorded details
- Benchmarks: FrontierSWE v2 得分 55.0%
- Benchmarks: CursorBench 52.8%，默认 effort
- Benchmarks: FrontierCode 1.1 得分 60.5%
- Benchmarks: Intelligence Index 约 60% 的 GPT-6 Astra 单任务成本
- Benchmarks: AA-Omniscience 幻觉率 15%
- Benchmarks: Arena High 排名第五
- Benchmarks: 单图文本定位/识别 84.7%，目标检测 85.1%
- Benchmarks: CursorBench 得分 52.8%
Evidence coverage
The KG currently records 14 known information dimensions and has 5 undocumented dimensions, linked to 10 public source articles. These counts describe record coverage, not model performance or verified fit.
Related public records
- Cursor — integrates
- Google Cloud — releases
- Google DeepMind — develops
- Google DeepMind — releases
- Vals AI — related to