About LMRank
LMRank is an independent guide to major AI models. Rank, pricing, capabilities, and benchmark context in one clear view, without the marketing noise.
Who runs LMRank
A side project I started in 2025 after too many evenings cross-referencing provider pages to find the best model for a task. It now tracks 120 models from 34 labs, maintained alongside my day job.
What LMRank covers
LLMs from major labs: OpenAI, Anthropic, Google DeepMind, Meta, Mistral, DeepSeek, and others. Organized by rank, provider, category, pricing, context window, and public benchmark evidence. Every model page shows per-token pricing and capability flags like vision, tool use, and reasoning.
How to use it
Start on the leaderboard for the broad market view, ranked by composite benchmark score. Drill into categories like coding, reasoning, or creative writing for shortlists. Model pages give pricing, context limits, release notes, and benchmark breakdowns. The comparison tool puts any two models side by side so you can weigh cost, speed, and capability.
How the site is maintained
Listings update within a day or two of any release, price change, or public benchmark. Every data point comes from official provider docs, published research, or public benchmarking platforms. The full methodology is on the Methodology page, and the feedback page lets you flag corrections or suggest models.