⌕Is ___ SOTA?
benchmarks / swe-bench-verified

SWE-bench Verified

resolved. Status display_only. Contamination risk high. Maintainer Epoch AI runs. Source.

Display only.This benchmark cannot decide a crown under methodology v1.0.
ModelScoreDateKind
GLM 5.20.7872026-06-16observation
Qwen3 70.77272727272727272026-05-19observation
DeepSeek V4 Pro0.77639751552795032026-04-24observation
Gemini 3.5 Flash0.79338842975206612026-05-19observation
Qwen3 6 Max Preview0.76652892561983472026-04-20observation
GLM 5.10.74172026-04-07observation
Qwen3 6 Plus0.57851239669421482026-03-31observation
Kimi K2 60.76652892561983472026-04-20observation
GPT-5.5 Pre Release0.80578512396694212026-04-23observation
Claude Opus 4.70.83471074380165292026-04-16observation
GPT-5.40.7685950413223142026-03-05observation
GPT-5.3 Codex0.74793388429752062026-02-05observation
Gemini 3.1 Pro Preview Customtools0.7561983471074382026-02-19observation
Claude Sonnet 4.60.75206611570247942026-02-17observation
Gemini 3 Flash Preview0.75413223140495872025-12-17observation
Claude Opus 4.60.78719008264462812026-02-05observation
GPT-5.10.67975206611570252025-11-13observation
Kimi K2 50.7376033057851242026-01-27observation
GLM 50.72077922077922072026-02-11observation
Gemini 2.5 Pro0.57556935817805392025-06-17observation
Gemini 3 Pro Preview0.72933884297520662025-11-18observation
GPT-5.20.7376033057851242025-12-11observation
O30.62318840579710142025-04-16observation
Claude Opus 4.10.73347107438016532025-08-05observation
GPT-4O0.309917355371900852024-11-20observation
GPT-4.10.485416666666666662025-04-14observation
GPT-50.73553719008264462025-08-07observation
Claude Opus 40.70661157024793392025-05-22observation
Claude Opus 4.50.76652892561983472025-11-24observation
Claude Sonnet 4.50.71280991735537192025-09-29observation