List your company

Software Issue Resolution on SWE-bench Lite (test)

185Resolved Issues CountCodex CLI GPT-5.4
151.72160.36169177.64May 7, 2026

Evaluation results

MethodLinks
Codex CLI GPT-5.4Config=default, Effort=high2026.05185236.73220.67
Codex CLI GPT-5.4Config=default, Effort=medium2026.05172161.82155.96
Spell GPT-5.4Config=default, Effort=medium2026.05171102.1251.37
Spell GPT-5.4Config=default, Effort=high2026.05171177.3849.26
Codex CLI GPT-5.4Config=default, Effort=low2026.05170136.79134.09
Spell GPT-5.4Config=default, Effort=low2026.0515368.9474.3