{"id":26138383,"url":"https://github.com/stacklok/pepsi-challenge","last_synced_at":"2026-06-08T08:32:02.727Z","repository":{"id":278569379,"uuid":"935625606","full_name":"stacklok/pepsi-challenge","owner":"stacklok","description":"LLM pepsi-challenge","archived":true,"fork":false,"pushed_at":"2025-08-06T13:28:00.000Z","size":647,"stargazers_count":1,"open_issues_count":16,"forks_count":1,"subscribers_count":8,"default_branch":"main","last_synced_at":"2026-06-05T05:32:42.526Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"TypeScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/stacklok.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-02-19T18:45:21.000Z","updated_at":"2026-05-29T13:32:00.000Z","dependencies_parsed_at":"2025-03-07T13:39:14.863Z","dependency_job_id":"d84da70d-48f1-4661-89ba-a1beea09c410","html_url":"https://github.com/stacklok/pepsi-challenge","commit_stats":null,"previous_names":["stackloklabs/pepsi-challenge","stacklok/pepsi-challenge"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/stacklok/pepsi-challenge","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/stacklok%2Fpepsi-challenge","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/stacklok%2Fpepsi-challenge/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/stacklok%2Fpepsi-challenge/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/stacklok%2Fpepsi-challenge/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/stacklok","download_url":"https://codeload.github.com/stacklok/pepsi-challenge/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/stacklok%2Fpepsi-challenge/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34055249,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-08T02:00:07.615Z","response_time":111,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-03-11T01:58:01.111Z","updated_at":"2026-06-08T08:32:02.709Z","avatar_url":"https://github.com/stacklok.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Pepsi-Challenge: LLM Model Evaluation\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"assets/pepsi-logo.png\" alt=\"Pepsi Taste Logo\" width=\"400\"/\u003e\n\u003c/p\u003e\n\n## Overview\n\nPepsi-Challenge is a server application for testing and validating Stacklok fine-tuned models. It provides support for both NVIDIA GPUs (CUDA) and Apple Silicon (Metal) platforms.\n\n### Hardware Support\n\n- **NVIDIA GPUs**: CUDA 12.1\n- **Apple Silicon**: Metal Performance Shaders (MPS)\n\n## Getting Started\n\n### Prerequisites\n\n- Python 3.12\n- Node.js v22.x (for frontend)\n- NVIDIA GPU with CUDA 12.1 or Apple Silicon device\n\n### Installation\n\n1. **Clone the repository**:\n```bash\ngit clone https://github.com/stacklok/pepsi-challenge.git\ncd pepsi-challenge\n```\n\n2. **Set up the backend**:\n```bash\ncd backend\n\n# Create and activate Python virtual environment\npython3.12 -m venv venv\nsource venv/bin/activate  # Unix/macOS\n# or\n.\\venv\\Scripts\\activate  # Windows\n\n# Install dependencies based on platform\n# For MacOS:\npip install -r requirements-macos.txt\n# For CUDA systems:\npip install -r requirements-cuda.txt\n```\n\n### Configuration\n\n1. **Backend Environment Setup** (`backend/.env`):\n```env\n# GitHub OAuth\nGITHUB_CLIENT_ID=your-github-client-id\nGITHUB_CLIENT_SECRET=your-github-client-secret\nGITHUB_CALLBACK_URL=https://your-domain.com/auth/callback\n\n# Model Configuration\nBASE_MODEL_NAME=Qwen/Qwen2.5-Coder-0.5B\nFINETUNED_MODEL_NAME=stacklok/Qwen2.5-Coder-0.5B-codegate\nFRONTEND_URL=https://your-domain.com\n\n# Access Control\nALLOWED_USERS=admin1,admin2\nADMIN_USERS=admin1,admin2\n\n# Development Settings\nWATCHFILES_FORCE_POLLING=false\nWATCHFILES_IGNORE_PATHS=*/unsloth_compiled_cache/*\n```\n\n2. **Frontend Environment Setup** (`frontend/.env`):\n```env\nNEXT_PUBLIC_BACKEND_URL=http://localhost:5000\nNEXT_PUBLIC_API_URL=http://localhost:5000\n```\n\n### GitHub OAuth Setup\n\n1. Create a new OAuth application at https://github.com/settings/developers\n2. Configure the application:\n   - Homepage URL: `https://your-domain.com`\n   - Authorization callback URL: `https://your-domain.com/auth/callback`\n3. Copy the client ID and secret to your backend `.env` file\n\n## Running the Application\n\n### Local Development\n\n1. **Start the Backend in Development Mode**:\n```bash\ncd backend\n# Enable hot-reload for development\nuvicorn main:app --host 127.0.0.1 --port 5000 --reload\n```\n\n2. **Start the Frontend in Development Mode**:\n```bash\ncd frontend\n# Install dependencies if not already done\nnpm ci\n# Start development server\nnpm run dev\n```\n\nThe application will be available at:\n- Frontend: http://localhost:3000\n- Backend API: http://localhost:5000\n\nFor local development, ensure your `.env` files are configured with local URLs:\n\n1. **Backend** (`backend/.env`):\n```env\nFRONTEND_URL=http://localhost:3000\nGITHUB_CALLBACK_URL=http://localhost:5000/auth/callback\nLOCAL_ENV=True\n```\n\n2. **Frontend** (`frontend/.env`):\n```env\nNEXT_PUBLIC_BACKEND_URL=http://localhost:5000\nNEXT_PUBLIC_API_URL=http://localhost:5000\n```\n\n### Production Deployment\n\nWe recommend building a Docker container for the frontend and backend,\nand then using that to run the application.  Note that the backend\nuses CUDA, so you'll need a *build* machine that has CUDA and a GPU\npassed through.  You'll also need to pass through a GPU to the backend\ncontainer; the frontend container is just a normal-non-GPU container.\n\n1. **Start the Backend**:\n```bash\ncd backend\nuvicorn main:app --host 127.0.0.1 --port 5000\n```\n\n2. **Start the Frontend**:\n```bash\ncd frontend\nnpm install --omit=dev\nnpm run build\nnpm run start\n```\n\n## Access Control\n\n- **Regular Users**: Configure allowed GitHub usernames in `ALLOWED_USERS`\n- **Administrators**: Configure admin GitHub usernames in `ADMIN_USERS`\n\nThese users will have access to the application and admin panel respectively.\n\n## Config Models\n\nAt the moment, the application supports running the comparison of the two models. These\nshould be configured in the `.env` file.\n\n```env\nFINETUNED_MODEL_NAME=stacklok/Qwen2.5-Coder-0.5B-codegate\nBASE_MODEL_NAME=Qwen/Qwen2.5-Coder-0.5B\n```\n\nThe application will then perform a comparison of the two models.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fstacklok%2Fpepsi-challenge","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fstacklok%2Fpepsi-challenge","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fstacklok%2Fpepsi-challenge/lists"}