Four-day head-to-head: Opus 5.5, GPT 6.1 Sol, Gemini 4 Argon, DeepSeek V4.1

CXCarroll @CXCarrollOriginally posted 2026.10.18

Open original on X

I have spent the last four days testing four models head to head. My test was on a common workflow for me building out a small, greenfield application. Models tested: 1. Opus 5.5 (CC) 2. GPT 6.1 Sol (Codex) 3. Gemini 4 Argon (Antigravity) 4. DeepSeek V4.1 Flash (OpenCode)