Skip to main content
Contributing to AI Stats is simple, transparent, and open to everyone.
This guide walks you through the entire process, whether you’re editing a model entry, updating benchmark data, improving documentation, or contributing new ideas.

​
Before you start

You’ll need a few basic tools installed:
ToolPurposeInstallation
GitVersion control to clone and push changes.Install Git
Node.js (v20+)Runs the AI Stats site locally (Next.js).Download Node.js
pnpmPackage manager used in this project.npm install -g pnpm
GitHub accountRequired for forks, issues, and pull requests.Sign up here
If you only want to edit data or Markdown documentation, you can make changes directly on GitHub without setting up a local environment.

​
1. Fork the repository

Go to the AI Stats GitHub Repository and click “Fork” in the top-right corner.
This creates your own copy of the project that you can freely edit.

​
2. Clone your fork

git clone https://github.com/yourusername/AI-Stats.git
cd AI-Stats
This downloads your fork to your local machine.

​
3. Create a new branch

It’s best practice to make a new branch for each change you make: 
git checkout -b update-benchmark-data
Please use a descriptive branch name that reflects your change to make reviewing your changes easier. Examples:
  • fix-{model-name}
  • add-new-organisation
  • improve-docs
  • update-pricing-for-{model-name}

​
4. Make your changes

Depending on what you want to edit:
AreaFilesDescription
Model data/apps/web/data/models/**/*.jsonAdd or update model metadata, pricing, or benchmarks.
Organisation data/apps/web/data/organisations/**/*.jsonUpdate details about companies or research groups.
Benchmark data/apps/web/data/benchmarks/**/*.jsonSubmit new evaluation results or corrections.
Documentation/apps/docs/v1/**/*.mdxImprove or expand guides, references, and docs pages.
Gateway/apps/api/*Make changes to the API Gateway.
If you’re editing data, please follow the schemas and field formats described in Repo Structure.

​
5. Preview your changes (optional)

If you’d like to preview your edits locally: 
pnpm install
pnpm dev
Use http://localhost:3100 for web changes, and http://localhost:3101 for docs changes. All data changes will have to be previewed after a PR has been made as the web app pulls from the database.

​
6. Commit your changes

Once you’re satisfied, commit your work: 
git add .
git commit -m "Updated benchmark data for Claude 3.5 Sonnet"
Please ensure your commit message is descriptive of the changes made to help reviewers understand your contribution.

​
7. Push your branch

git push origin update-benchmark-data

​
8. Create a Pull Request (PR)

  1. Go to your fork on GitHub.
  2. You’ll see a message offering to create a PR - click “Compare & Pull Request.”
  3. Fill out the PR template with the appropriate information for your changes
  4. Submit the PR!
Once your PR is reviewed and approved, it will be merged into the main AI Stats repository.

​
9. Celebrate 🎉

Your contribution will be live once merged. If you made it this far, we appreciate your effort and support and will be forever grateful for all your help! ❤️

View Contributors

See everyone who’s helped build AI Stats.

​
Contribution tips

  • Keep PRs small and focused - one logical change per PR.
  • Attempt to follow conventional commits where possible. Use clear commit prefixes like fix:, add:, or chore:.
  • If your change affects data accuracy, include sources or references.
  • For large or uncertain ideas, open a GitHub Discussion or message in Discord before coding.
  • Always stay respectful and follow our Code of Conduct.

​
Example workflow summary

# 1. Fork and clone

git clone https://github.com/yourusername/AI-Stats.git
cd AI-Stats

# 2. Create branch
git checkout -b add-openai-new-model

# 3. Make changes
# e.g., update /data/models/new-openai-model.json

# 4. Commit and push
git add .
git commit -m "Add new OpenAI model entry"
git push origin add-new-openai-model

# 5. Create a Pull Request on GitHub

​
Next steps

Repo Structure

Learn about the folder layout and where to find key data files.

Editing a Model

Start with a practical example — updating or adding a model entry.