DeepSeek V3.2
by DeepSeek
70.0
best score
70.0%
Best Score
bash-only
Best Leaderboard
1
Models Used
Yes
Open Source
Score History
| Entry | Score |
|---|---|
| DeepSeek V3.2 (high reasoning) | 70.0% |
| mini-SWE-agent + DeepSeek V3.2 (high reasoning) | 70.0% |
| DeepSeek V3.2 | 59.0% |
| DeepSeek V3.2 Reasoner | 60.0% |
| mini-SWE-agent + DeepSeek V3.2 Reasoner | 60.0% |