o3-2025-04-16· OpenAI LMArena Human Preference 140KWhat is 9.11 - 9.9 Save Share Download ReportFrom a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with claude-3-5-haiku-20241022’s answer. UserCopy promptWhat is 9.11 - 9.9 AssistantCopy response9.11 − 9.9 = −0.79Copy first promptCopy first responseOther models, same or similar promptResponses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.Identical first promptclaude-3-5-haiku-20241022Compare side by side →