Share your thoughts, 1 month free Claude Pro on usSee more

Software Engineering on SWE-Bench Multilingual 1.0 (test)

75.2Resolution Rate

Claude-Opus-4.5

Updated 1mo ago

Evaluation Results

Method	Links
Claude-Opus-4.5 2026.02		75.2
Claude-Opus-4.5 2026.02		71.8
Claude-Opus-4.5 2026.02		71.7
MiniMax-M2.1 2026.02		67.5
Claude-Sonnet-4.5 2026.02		67.2
MiniMax-M2.1 2026.02		66.2
Claude-Sonnet-4.5 2026.02		66
Qwen3-Coder-Next 2026.02		64.3
GLM-4.7 2026.02		63.7
Kimi-K2.5 2026.02		63.7
Qwen3-Coder-Next 2026.02		62.8
MiniMax-M2.1 2026.02		62.5
DeepSeek-V3.2 2026.02		62.3
Kimi-K2.5 2026.02		62.3
DeepSeek-V3.2 2026.02		61.8
GLM-4.7 2026.02		61.8
Claude-Sonnet-4.5 2026.02		61.3
GLM-4.7 2026.02		60.8
Qwen3-Coder-Next 2026.02		56.2
DeepSeek-V3.2 2026.02		55.5