ai.OpenAI
GPT-5.2 Thinking
Pretty great at coding, abstract reasoning, science, math, long context.
LMArena / Text Arena
not reported
LMArena / Text Arena
pretty greatpretty okaynot yetTap a verdict for the number and source.
Measures which of two anonymous chat answers people prefer.
Why it matters
Captures preferences that right-answer tests miss.
How to read it
Elo-style preference rating. Uncertainty intervals can overlap. Higher is better.
Run by Arena. Last checked 2026-10-14.
Visit the benchmark source ↗MODEL / VENDORLMArena / Text Arenaconversation
OpenAI
GPT-5.1 Thinking
Pretty great at coding, science.
LMArena / Text Arena
not reported
Anthropic
Claude Opus 4.6
Pretty great at coding, terminal work.
LMArena / Text Arena
pretty okay
1498 Elo
Text Arena overall, no adjustments. Leaderboard dated October 13, 2026. ±3.
Source ↗Observed 2026-10-14Google
Gemini 4 Argon High
Pretty great at conversation.
LMArena / Text Arena
pretty great
1525 Elo
Text Arena overall, no adjustments. Leaderboard dated October 13, 2026. Preliminary, ±9.
Source ↗Observed 2026-10-14The top score means the top sourced score in this snapshot. Different harnesses, budgets and samples can change results. A blank cell is a blank cell.