A long-horizon software engineering benchmark from Datacurve with 113 original tasks across 91 active open-source repositories and five programming languages. It emphasizes multi-file changes, behavioral correctness, and reduced contamination risk.
Browse the latest scores, model modes, release dates, and parameter sizes for DeepSWE.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
| Rank | Model | License | |||
|---|---|---|---|---|---|
![]() GPT-5.6 Sol Thinking Level · Extra HighTools | 72.70 | 2026-06-26 | Unknown | Closed | |
![]() Claude Fable 5 Deep Thinking ModeTools | 70.00 | 2026-06-09 | Unknown | Closed | |
![]() GPT-5.6 Terra Thinking Level · Extra HighTools | 69.60 | 2026-06-26 | Unknown | Closed | |
4 | ![]() Claude Opus 5 Thinking Level · MaxTools | 68.80 | 2026-07-24 | Unknown | Closed |
5 | ![]() Kimi K3 Thinking Level · MaxTools | 67.50 | 2026-07-16 | 2800B | Conditional |
6 | ![]() GPT-5.6 Luna Thinking Level · Extra HighTools | 67.20 | 2026-06-26 | Unknown | Closed |
7 | ![]() GPT-5.5 Thinking Level · Extra HighTools | 67.00 | 2026-04-23 | Unknown | Closed |
8 | ![]() GLM-5.3 Thinking Level · MaxTools | 66.90 | 2026-08-14 | 744B | Conditional |
9 | Grok 4.6 Thinking Level · HighTools | 65.90 | 2026-08-12 | Unknown | Closed |
10 | ![]() Gemini 3.7 Flash Thinking Level · HighTools | 65.30 | 2026-08-13 | Unknown | Closed |
11 | ![]() Hy4 preview Thinking Level · HighTools | 64.30 | 2026-08-28 | 770B | Free Commercial |
12 | ![]() GLM-5.3-Flash Thinking Level · MaxTools | 63.40 | 2026-08-26 | 320B | Free Commercial |
13 | ![]() DeepSeek-V4-Pro Thinking Level · Extra HighTools | 62.70 | 2026-08-13 | 1600B | Free Commercial |
14 | ![]() DeepSeek-V4-Flash-Vision-Exp Thinking Level · MaxTools | 59.30 | 2026-08-21 | 305B | Free Commercial |
15 | ![]() Muse Spark 1.2 Thinking Level · Extra HighTools | 59.30 | 2026-08-05 | Unknown | Closed |
16 | ![]() Claude Opus 4.8 Deep Thinking ModeTools | 59.00 | 2026-05-28 | Unknown | Closed |
17 | ![]() Qwen3.8-Flash-Next Thinking Level · Extra HighTools | 58.70 | 2026-08-26 | 125B | Conditional |
18 | ![]() Qwen3.8-Max Thinking Level · Extra HighTools | 56.60 | 2026-08-03 | 2400B | Conditional |
19 | ![]() DeepSeek-V4-Flash Thinking Level · MaxTools | 54.40 | 2026-04-24 | 284B | Free Commercial |
20 | ![]() Claude Sonnet 5 Deep Thinking ModeTools | 54.00 | 2026-06-30 | Unknown | Closed |
21 | ![]() Muse Spark 1.1 Thinking EnabledTools | 53.30 | 2026-07-09 | Unknown | Closed |
22 | Grok 4.5 Thinking Level · HighTools | 53.00 | 2026-07-08 | Unknown | Closed |
23 | ![]() GPT-5.4 Thinking Level · Extra HighTools | 52.00 | 2026-03-05 | Unknown | Closed |
24 | ![]() Gemini 3.6 Flash Thinking EnabledTools | 49.00 | 2026-07-21 | Unknown | Closed |
25 | ![]() GLM-5.2 Deep Thinking ModeTools | 44.00 | 2026-06-13 | 753.3B | Free Commercial |
26 | ![]() Qwen3.8-27B Thinking EnabledTools | 42.20 | 2026-08-14 | 27B | Free Commercial |
27 | ![]() Gemini 3.5 Flash Thinking Level · MediumTools | 37.00 | 2026-06-20 | Unknown | Closed |
28 | ![]() Kimi K2.7 Code Standard ModeTools | 31.00 | 2026-06-12 | 1000B | Free Commercial |
29 | ![]() Claude Sonnet 4.6 Thinking Level · HighTools | 30.00 | 2026-02-17 | Unknown | Closed |
30 | ![]() Hy3 Thinking Level · HighTools | 28.00 | 2026-07-06 | 295B | Free Commercial |