List your company

Software Engineering Issue Resolution on SWE-Bench Multilingual

45.78Resolve RateNemotron 3 Super
-1.612810.691122.99535.2989Dec 26, 2025Jan 13, 2026Jan 31, 2026Feb 18, 2026Mar 8, 2026Mar 26, 2026Apr 14, 2026

Evaluation results

MethodLinks
Nemotron 3 SuperFramework/Scaffold=OpenHands2026.0445.78
GPT-OSS-120BFramework/Scaffold=OpenHands2026.0430.8
HybridFeedback=Hybrid (Execution-free + Execution-based)2025.120.357
Execution-based onlyFeedback=Execution-based2025.120.333
Execution-free onlyFeedback=Execution-free (SWE-RM)2025.120.33
Poor Calibrated RMFeedback=Ablated RM (Poorly Calibrated)2025.120.21