Llmbenchmark
Which LLM for which task: benchmark scores and API prices for 400+ models. Free, no key.
It draws on 222 public benchmarks and the API prices of 386 LLMs. It is free, needs no key and allows 500 requests per IP per day. The server speaks MCP over Streamable HTTP.
The data is licensed CC BY 4.0. It can recommend models for a given task with price, context or open-weights limits. It provides leaderboards for categories including overall, coding, agentic, reasoning, math, knowledge, or design.
Data refreshes once a day. No language model is involved; the same text and data always return the same answer.
Summary drafted from the project's own website. Every sentence is backed by text on that page and was reviewed before publishing.