Battles
Every battle is a real run: the published task suite for that category, both contestants, a panel of blind position-swapped judges, raw outputs downloadable. No battle, no verdict.
Writing · tested 7 Aug 2026
ClaudevsChatGPT
5–4
Too close
Everyday chat · tested 7 Aug 2026
ChatGPTvsGemini
3–1
Too close
Images · tested 17 Jul 2026
GeminivsGPT Image (in ChatGPT)
2–4
Leans · not proven yet
If you write code
These compare things you reach through an API key, and the tasks behind them are programming tasks. More for developers →
Best free · tested 7 Aug 2026
Gemini 3.5 FlashvsDeepSeek V4 Flash
9–2
Leans · not proven yet
Best-value API · tested 7 Aug 2026
DeepSeek V4 ProvsGPT-5.6 Terra
4–7
Too close
Best-value API · tested 7 Aug 2026
GLM 5.2vsClaude Opus 4.6
8–3
Leans · not proven yet
Coding · tested 7 Aug 2026
CursorvsGitHub Copilot
1–4
Leans · not proven yet