Models Pie – LLM Model Comparison & Ranking Tool

A simple web tool that ranks LLM models based on your custom trade-off priorities (fast vs. cheap vs. good quality). Users drag sliders on a pie chart to weight their preferences, and a live-updating table shows ranked models from BenchLM, OpenRouter, and Artificial Analysis. Solves decision paralysis when choosing between dozens of models.

The LLM model selection problem is genuinely acute right now — there are over 100 models available via OpenRouter alone, and the landscape changes weekly, which creates real friction for developers trying to make defensible technical choices. Artificial Analysis already does benchmarking, and LMSYS Chatbot Arena handles quality rankings, but neither lets you weight trade-offs interactively, so there's no clear incumbent for the preference-weighted comparison angle. The $500–2k/mo ceiling makes sense only if monetization is via a Pro tier or API access, since the core tool will attract developers who expect free tools and won't convert easily — ad revenue on a niche devtool is essentially zero. The biggest risk is that this is a dashboard, not a workflow, meaning it solves a one-time decision rather than a recurring need, which makes retention nearly impossible and caps growth to whatever organic search volume the model-comparison query space can deliver.