Early version. All scores are real results. The arena and automatic scoring are planned; for now a maintainer runs each evaluation.
Open Buddhist AI benchmarks and skills

Find the best AI for every Buddhist task.

We test AI models and skills on real Buddhist work and share all the results for free. For each task, see the top performance (the highest score) and the best value (a high score for much less money). If you make AI models or skills, send us yours and we will test them too.

For Buddhist projects

Find the best AI for your task

For AI developers

Share your solution, join the evals

Skills, models, workflows or test data. Paste a link and we score it on the open benchmarks for free. Lead a task and we announce it.

No repo? Share your files: a skill, project instructions, a Claude file, a style guide or a prompt. Needs a free GitHub account (make one). The form is public.

15tasks defined
1with a live benchmark
$0to get scored
Skills

Skills by task

Grouped by the kind of work. Each task shows two picks: the highest score, and the best score at a fifth of the cost or less.

Install the best skills

One plugin per language holds the top skill for every task. Pick the best-value version if cost matters more than the last few points.

Pick your language. One plugin holds the top skill for every task.
1. Add the BuddhistAI marketplace
/plugin marketplace add BuddhistAI/buddhist-skills
2. Install the plugin
/plugin install buddhist-skills-tibetan@buddhistai
Then ask: "Extract the sa-bcad outline of this commentary."

Prepare texts

From scan or e-text to a clean, structured source.

Table of contents

Measured
ས་བཅད།

Finds the sa-bcad outline headings in a commentary, with their level and position.

Top performance0.962sa-bcad method · Sonnet 5.5 · ≈$1.20 / 100 pages
Best value0.824GEPA-optimized · Haiku 5.5 · ≈$0.05 / 100 pages
toc-extraction-tibetanLeaderboardTask repo

Study and annotate

Find who, where and what a text cites.

Text-commentary alignment

Needs a benchmark

Link each root-text passage to the commentary that explains it.

Translate

Glossary, draft, alignment and review.

Term extraction

Needs a benchmark

Propose glossary terms and renderings from a source and its translation.

Translation

Arena planned

Draft translation. Scored by side-by-side reader votes as well as automatic metrics.

Translation review

Needs a benchmark

Check a translation against its source, glossary and style guide.

Art, practice and teaching

Shared by the community, not yet measured.

Sharing needs a free GitHub account (make one). The form is public.

Leaderboard

Which models and skills win each task

All tasks on one page. Models are ranked on every benchmark with no skill, including Q&A benchmarks. Skills are ranked on top of each model; the top skill is the one in buddhist-skills, and a challenger that beats it replaces it.

New leader · 2026-11-13TOC extraction (sa-bcad method)Table of contents, Tibetan · 0.962 on Sonnet 5.5
New best value · 2026-11-13TOC extraction (GEPA-optimized)Table of contents, Tibetan · 0.824 for $0.05 / 100 pages
Next leaderYour model or skill herePaste your repo and we announce the winner on LinkedIn and X.
Task
Model
Show
#SkillModelScoreGainCost / 100 pages
85 passages from two commentaries, one book held out. Last run 2026-11-13. Full live leaderboard
Arena · planned

Read two answers. Pick the better one.

Some work has no single right answer. Translators and teachers vote blind between two answers, from a model on its own or a model with a skill, and the votes become a ranking for each task and language.

Source · Thirty-Seven Practices, verse 1 · Translate into English
དལ་འབྱོར་གྲུ་ཆེན་རྙེད་དཀའ་ཐོབ་པའི་དུས། །རང་གཞན་འཁོར་བའི་མཚོ་ལས་བསྒྲལ་བྱའི་ཕྱིར། །
Answer ANow that we have this great ship of freedom and fortune, so hard to find, to carry ourselves and others across the ocean of samsara...
Answer BAt this time, having obtained the great boat of leisure and endowment, difficult to find, in order to liberate self and others from the sea of cyclic existence...
The skills and models stay hidden until you vote.

Translation · Tibetan to English

Example layout. No votes yet.
#EntryRatingVotes
1Skill A + model X–0
2Skill B + model Y–0
3Model X, no skill–0
Vote on answers
Request

Need AI for a task we don't cover yet?

Tell us what you need to do. Every request goes on the public Skills board. People can vote for it, and developers can pick it up. The most wanted tasks get a benchmark first.

Send request on GitHub
This opens a short form on GitHub. You need a free GitHub account.
Not sure yet? Ask in our community discussions under "Find the best AI" or "Requests and ideas".

Most wanted

Requests and their votes are on the public Skills board. Add yours, or vote for the tasks you need.

Benchmarks

The test sets behind the scores

Open and collaborative. Each benchmark lives in its own repo with a maturity level, so you know how much to trust a score. Groups may publish competing sets; new skills are scored on all of them.

1 · ExampleFormat check only
2 · DraftReal items, one author
3 · ReviewedChecked by a second reader
4 · StableHeld-out set, used in practice
BenchmarkRanksLanguageItemsMaturityMaintainer
TOC extraction 2.0.0SkillsTibetan85 Draft, answer key needs a reader checkBuddhistAI
Thai Nak Tham examsModelsThai4,075 Reviewed, public, may be memorizedImported
Tibetan OCROCR enginesTibetan– Stable, run by BDRCBDRC
BuddhistBench Q&A, by traditionModelsMany– Example items, 15 tracks openBuddhistAI

Have exam papers or a test set your team already uses? Paste its repo to publish it as a benchmark.

Workflows

How Buddhist projects use AI

Short guides that chain tasks together, written with the teams who do this work. Green steps have a measured skill today.

Prepare a commentary for translation

text-cleanup→toc-extraction→segmentation→text-commentary-alignment

From an OCR file to a structured, segmented source with its outline.

Build a project glossary

text-translation-alignment→term-extraction→entity-extraction

Collect renderings from past translations into one reviewed glossary.

Review a draft translation

translation→citation-mapping→translation-review

Check a draft against source, glossary and style guide before an editor sees it.

Looking for other Buddhist AI tools or data? The Directory lists Buddhist AI projects and datasets from all traditions and languages. Add yours. To choose an AI for translating into a Buddhist language, see Languages: data, AI projects and translation quality for each language, and an overview of Buddhist countries.

Contribute

Paste a link. We test it for you.

Skills and benchmarks here are open source and built together. Anyone can add a skill, a model, a test set, or a fix to any of them. Paste a link to get started; evaluation is free, paid by a sponsor.

Today your link opens a GitHub issue and a maintainer runs the evaluation. Automatic checks are coming.

A folder with a SKILL.md is enough. Any language, any model.

No GitHub repo? Share a skill or any file you give to AI.

The prompt library holds files people give to AI: your CLAUDE.md, project instructions, a style guide, a glossary or a prompt. Drag your files into a short form. You need a free GitHub account, and the form is public. Your name and picture are optional.

Share your files

Made something? Show it in the community (Skills category).

    Improve a skill

    Every skill's task repo keeps all its variants. Fork the top one, change it, paste your fork. If it scores higher, it becomes the main skill.

    Grow a benchmark

    Add the edge cases your team runs into, fix a wrong answer, or publish your own set. Competing sets from different groups are all kept.

    Judge in the arena

    Read two answers and pick the better one. Votes from translators and teachers rank the open-ended tasks.

    Review a benchmark

    Know the material well? Check a benchmark's questions and answers in your field. You are named as reviewer, and the benchmark moves up to Reviewed. Offer a review

    Donate compute or API credits

    Every score costs money to make. Credits pay for evaluations of everyone's submissions and never change a score or a ranking. Offer credits

    Learn by building

    Short tutorials: run the scorer, evaluate your first model on a Buddhist task, write your first skill, add a benchmark. Start learning

    For builders

    Where everything lives

    The homepage reads from five kinds of sources. Each surface above is built from them, so nothing on the page is typed in by hand.

    Workflow and task docsOne repo per taskbuddhist-skillsbuddhist-benchmarksOne repo per benchmark what each task is, how they chainall variants, ranked on its benchmarkstop-performing + best-value skill per taskindex of every benchmark per taskitems, scorer, maturity level SkillsLeaderboardArena install, two picks per taskdecides what ships to buddhist-skillsblind votes for open-ended tasks

    Contributors

    ngawangtrinley drupchen kaldan007 WeBuddhist SM

    How to contribute

    Supported by

    Unione Buddhista Italiana Buddha Educational Foundation OpenPecha

    Support evaluations