LLM Benchmark (llmbenchmark.ai) is a free LLM leaderboard that collects public LLM benchmarks in one place for developers and teams who pick a model for real work. It ranks models overall and by category: coding, agentic work, reasoning, math,
...
knowledge and web design, so you can compare LLMs on the AI benchmarks that matter for your task. Every score is read from the public board that published it and links back to it, with the date it was read. The site also shows API list prices, the cost of one solved task and head-to-head comparisons of any two models. Developers can use the same data for free through a REST API and an MCP server for Claude Code, Cursor and Codex, with no key needed. LLM Benchmark launched in October 2026 and is run by Take Group sp. z o.o., a software company from Katowice, Poland.
