Public KG record
DeepSeek-V4.1-Flash-1017
Recorded overview
Reaches 57.0% pass@1 on CWE-bench v2 at $0.04 per rollout.
An evidence-backed entity from the project knowledge graph; detailed tool metadata is still being added.
Recorded details
- Benchmarks: DeepSWE v1.1：77.0
- Benchmarks: HLE（无工具）：42.6
- Benchmarks: Terminal-Bench 4.0：42.3
- Benchmarks: DeepSWE v1.1 得分 77.0
- Benchmarks: DeepSWE得分77.0
- Benchmarks: DeepSWE v1.1得分77.0
- Benchmarks: 57.0% pass@1 on CWE-bench v2
- Benchmarks: NL2Repo-Bench 68.2
Evidence coverage
The KG currently records 12 known information dimensions and has 7 undocumented dimensions, linked to 6 public source articles. These counts describe record coverage, not model performance or verified fit.
Related public records
- DeepSeek — releases
- Claude Opus 5 — competes with
- DeepSeek-V4.1-Flash — based on
- Claude Opus 5 — competes with
- Inkling 2 Small — competes with