Benchmarks

Benchmarks by model

29 models with published scores on standard tests. Each page shows the model's category profile, its rank on every test it reports, and the source for each number.

OpenLLM Buddy catalog

Open-weight models you can run on a dedicated GPU through OpenLLM Buddy today.

Open weights

Other open-weight releases, for reference. Weights are public; most need large multi-GPU servers.

Closed API

Frontier models available only through their lab's API.

Scores are published results from model cards, lab reports, and public leaderboards, each linked to its source. Settings such as thinking mode or test version differ between sources and are shown next to each score. OpenLLM Buddy catalog models are uncensored fine-tunes; their scores are those of the official base model. Data collected 2026-10-30. How the index works