Benchmark
DeepSWE Benchmark
What is DeepSWE?
Mini-swe-agent harness, temperature=0.95, top_p=1.0, max_new_tokens=64k under 1M context. Benchmark Radar tracks 3 reported scores for DeepSWE, cited by 3…
- Released
- 2025-09-15
- Evaluates
- coding agent
- Openness
- unknown
- Importer
- Model reports
- Reported scores
- 3
Source provenance
- Original evidence https://github.com/deepswe/deepswe
DeepSWE code
DeepSWE results and reported scores
Model reports
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| DeepSeek-V4.1-Flash | DeepSeek | 74.2 | 2026-09-10 | evidence |
| GLM-5.3-Flash | Z.ai | 63.4 | 2026-08-27 | evidence |
| Hy4 preview | Tencent | 64.3 | 2026-08-28 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.
Which sources cite DeepSWE?
- DeepSeek-V4.1-Flash DeepSeek, 2026-09-10
- Hy4 preview Tencent, 2026-08-28
- GLM-5.3-Flash Z.ai, 2026-08-27