LLMatcher

Hey Product Hunt! 👋 I'm Paweł, and I built LLMatcher in a single 10-hour session this week. Here's the story: The Problem I Had: I kept switching between ChatGPT, Claude, and Gemini depending on the task, but I never knew which one was actually better for MY specific needs. Benchmarks are great, but they don't tell you which model works best for how YOU work. The Solution: Blind A/B testing. You see two anonymous AI responses to your prompt, vote for the better one, and after 50 votes (~15 min), you unlock personalized recommendations based on YOUR preferences. The Pivot: Initially positioned it as "help us rank AI models" (community contribution). Crickets. Repositioned to "find YOUR perfect AI in 50 votes" (personal utility). Way better response. Lesson learned: people care about solving their own problems first. Current Status: - Live and working - 6 flagship models: GPT-5.2, Claude Opus 4.5, Claude Sonnet 4.5, Gemini 3 Pro, Grok 4, Grok 4.1 Fast - Free, no signup required - Will add more models based on demand What I Need From You: 1. Try it! Vote on a few prompts and tell me if the experience makes sense 2. Feedback on the 50-vote barrier - too high? too low? 3. What models or categories would you want to see? Thanks for checking it out. Happy to answer any questions! 🚀