Harness comparisons

2 posts from X ran Google Antigravity and OpenAI Codex on the same task. Each side used its own model, so harness and model effects are mixed. Models: Claude Opus 5.5, GPT-6.1 Sol, Gemini 4 Argon, Gemini 3.1 Pro, DeepSeek V4 Pro 0813, GPT-6 Sol. All harness comparisons.

Claude Opus 5.5GPT-6.1 SolGemini 4 Argon

Gemini 3.1 ProDeepSeek V4 Pro 0813GPT-6 Sol

No matches. Submit a comparison