SOTA Software Engineering benchmarks and papers with code

Benchmarks

Dataset Name	SOTA Method	Metric
SWE-bench Verified	JoyAI-LLM Flash	Accuracy62.6	33	13d ago
SWE-bench Verified	OpenHands	Success Rate71.8	29	4d ago
SWE-bench Verified	MemCoder	Resolution Rate83.8	26	18d ago
SWE-Bench Multilingual 1.0 (test)		Resolution Rate75.2	20	1mo ago
SWE-Bench Verified	Mini-SWE	Pass@184	18	3d ago
SWE Verified		Resolution Rate77.2	17	1mo ago
SWE-Bench Pro 1.0 (test)		Resolved Rate51.6	14	1mo ago
SWE-bench Lite	DFlash+DDTree	Speedup4.38	12	3d ago
SWE-bench	Qwen3.5-122B-A10B	Resolve Rate67.4	11	4d ago
Commit0-Lite	Claude Sonnet 4.5	Score59.5	9	25d ago
PaperBench	Claude Sonnet 4.5	Score66.8	9	25d ago
SWE-Bench Pro (public)	CCA	Resolve Rate (Pass@1)59	9	1mo ago
SWE-rebench January 2026 (test)		Resolved Rate52.9	8	1mo ago
SWE-bench Lite	TOOLSELF	Accuracy16.1	8	1mo ago
SWE-Bench Verified		SWE-Agent Score78.2	7	1mo ago
SWE-Bench (val)		Acc28.8	7	1mo ago
SWE-Bench Verified	ProRL Agent-14B (RL)	Reproduced Success Rate23.6	6	29d ago
SWE-bench Verified	GEA	Worst-Case Success Rate71	6	1mo ago
SWE-bench Verified		Generic Score72.4	4	17d ago
SWE-bench Verified 500 issues (Protocol A)	OpenHands CodeAct + GBT-SE	Success Rate (SR)73.6	4	1mo ago
SWE-Bench AgentLess Repair	MiMo-V2-Flash Base	Resolved Percentage30.8	4	1mo ago
SWE-Bench Verified	AgentSPEX	Score77.1	3	3d ago
SWE-Bench Python subset Pro	SageAgent	Resolution Rate59	3	1mo ago
SWE Verified MEDIUM reasoning	HarmonyAgent	Overall Score53.3	2	17d ago
SWE Verified HIGH reasoning	gpt-oss-20b	Accuracy60.7	2	17d ago

Showing 25 of 29 rows