List your company

Issue Resolution on SWE-bench Verified (test)

77.2Pass RateClaude-4.5-Sonnet
-1.746418.749339.24559.7407Dec 26, 2025Dec 31, 2025Jan 6, 2026Jan 12, 2026Jan 17, 2026Jan 23, 2026Jan 29, 2026

Evaluation results

MethodLinks
Claude-4.5-SonnetAgent Framework=ReAct Agent, Model Size=100B+, Scaffold=OpenHands2025.1277.2————
GPT-5.1Agent Framework=ReAct Agent, Model Size=100B+, Scaffold=OpenHands2025.1276.3————
Gemini-3-ProAgent Framework=ReAct Agent, Model Size=100B+, Scaffold=OpenHands2025.1276.2————
Qwen3-Coder-480B-A35BAgent Framework=ReAct Agent, Model Size=35/480B, Scaffold=OpenHands2025.1269.6————
Kimi-K2 (Sub-Agent)Localization Mode=Sub-Agent, Localization Model=FuseSearch-4B2026.0168.731.9290713,000—
Kimi-K2 (No Localization)Localization Mode=None2026.0168.441.13121,053,000—
Kimi-K2 (Pre-Search)Localization Mode=Pre-Search, Localization Model=FuseSearch-4B2026.0168.131.6223562,000—
DeepSeek-V3.1Agent Framework=ReAct Agent, Model Size=37/671B, Scaffold=OpenHands2025.1261————
Gemini-2.5-ProAgent Framework=ReAct Agent, Model Size=100B+, Scaffold=OpenHands2025.1259.6————
GLM-4.5-AirAgent Framework=ReAct Agent, Model Size=12/106B, Scaffold=OpenHands2025.1257.6————
SWE-CompressorAgent Framework=Folding Agent, Model Size=32B, Scaffold=OpenHands2025.1257.6————
Seed-OSS-36BAgent Framework=ReAct Agent, Model Size=36B, Scaffold=OpenHands2025.1255.2————
FrogBoss-32BAgent Framework=ReAct Agent, Model Size=32B, Scaffold=OpenHands2025.1254.6————
Threshold-Compression AgentAgent Framework=Summary Agent, Model Size=32B, Scaffold=OpenHands2025.1253.8————
Claude-3.5-SonnetAgent Framework=ReAct Agent, Model Size=100B+, Scaffold=OpenHands2025.1253————
SWE-Mirror-LM-32BAgent Framework=ReAct Agent, Model Size=32B, Scaffold=OpenHands2025.1252.2————
ReAct AgentAgent Framework=Summary Agent, Model Size=32B, Scaffold=OpenHands2025.1249.8————
DeepSeek-R1-0528Agent Framework=ReAct Agent, Model Size=37/671B, Scaffold=OpenHands2025.1245.6————
DeepSWE-32B-PreviewAgent Framework=ReAct Agent, Model Size=32B, Scaffold=OpenHands2025.1242.2————
Llama3-SWE-RL-70BAgent Framework=ReAct Agent, Model Size=70B, Scaffold=OpenHands2025.1241————
SWE-agent-LM-32BAgent Framework=ReAct Agent, Model Size=32B, Scaffold=SWE-agent2025.1240.2————
GPT-4oAgent Framework=ReAct Agent, Model Size=100B+, Scaffold=Agentless2025.1238.8————
R2E-Gym-32BAgent Framework=ReAct Agent, Model Size=32B, Scaffold=OpenHands2025.1234.4————
Qwen3-235B-A22BAgent Framework=ReAct Agent, Model Size=22/235B, Scaffold=OpenHands2025.1234.4————
SWE-Fixer-72BAgent Framework=ReAct Agent, Model Size=72B, Scaffold=SWE-Fixer2025.1232.8————
Lingma-SWE-GPT-72BAgent Framework=ReAct Agent, Model Size=72B, Scaffold=SWE-SynInfer2025.1228.8————
CWM (Meta)Size=32B, SWE Training Data=3M trajs from 3.15K repos2026.0122.22———15.88
GPT-4.1-miniSize=[10, 100]B, SWE Training Data=-2026.0121.79———17.85
SWE-Gym-32BAgent Framework=ReAct Agent, Model Size=32B, Scaffold=OpenHands2025.1220.6————
SWE-SPOTSize=4B, SWE Training Data=8K trajs in one target repo2026.0119.34———17.12
Mini-Coder-4BSize=4B, SWE Training Data=400K trajs from 128 repos2026.0118.76———8.7
Qwen3-Coder-30BSize=32B, SWE Training Data=-2026.0116.74———11.56
Gemini-2.5-Flash-LiteSize=[0.1, 10]B, SWE Training Data=-2026.018.8———7.46
GPT-5-nanoSize=[0.1, 10]B, SWE Training Data=-2026.016.49———2.97
Qwen3-4B-Instruct-2507Size=4B, SWE Training Data=-2026.013.32———3.2
Gemma-3-27b-itSize=27B, SWE Training Data=-2026.011.29———1.34