BuddhistBench
A question-and-answer benchmark of Buddhist knowledge across traditions. It ranks models (Claude, GPT, Gemini and others), with scores per tradition, and tests whether a model answers from inside the tradition a question names.
Coming soon
The benchmark is being built in BuddhistAI/buddhist-bench. Its leaderboard will appear here, grouped by track, language and mode (closed-book or with web search), as soon as the first runs are published. Right now no runs have been published yet.
To contribute questions, see the repo's CONTRIBUTING.md.

BuddhistAI