List your company

Coding Agent Capacity on SWE-Evo

27.1ScoreClaude-Opus-4.5
7.75612.77817.822.822Jun 30, 2026

Evaluation results

MethodLinks
Claude-Opus-4.52026.0627.1
Claude-Sonnet-4.52026.0616.7
Gemini-3-Flash High2026.0612.8
GPT-5.2 High2026.0612.5
Seed2.0 Lite2026.0610.6
GPT-5-mini High2026.0610.4
Gemini-3-pro High2026.068.9
Seed2.0 Pro2026.068.5