Awesome-LLM-Eval

Why we included this project

Choosing how to evaluate a large language model is often harder than picking the model itself, and this curated index is a good place to start. It gathers evaluation tools, benchmark datasets, leaderboards, and papers in one spot, sorted by what you actually want to test: general knowledge, coding, RAG pipelines, agent capabilities, long-context handling, multimodal behavior, and inference speed. The list also covers the models themselves and the frameworks used to train them, so you can see the full range of evaluation work in a single pass. Because the project is maintained alongside an academic survey on LLM evaluation, the selection reflects a considered view of the field rather than a random scrape of links. Teams planning an evaluation strategy can use it to find out which benchmarks and harnesses exist before committing to a specific toolchain.