{"id":18430409,"url":"https://github.com/aryanvbw/private-ai","last_synced_at":"2025-04-07T17:33:34.659Z","repository":{"id":211361136,"uuid":"728921096","full_name":"AryanVBW/Private-Ai","owner":"AryanVBW","description":"Private-AI is an innovative AI project designed for asking questions about your documents using powerful Large Language Models (LLMs). The unique feature? It works offline, ensuring 100% privacy with no data leaving your environment","archived":false,"fork":false,"pushed_at":"2024-02-26T22:02:25.000Z","size":12329,"stargazers_count":18,"open_issues_count":16,"forks_count":6,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-03-22T22:05:17.708Z","etag":null,"topics":["aryanvbw","chatgpt","chatgpt-4","chatgpt3","chatgpt4","gpt4","gpt4all","privategpt","vivek","vivek-w"],"latest_commit_sha":null,"homepage":"https://aryanvbw.github.io/Private-Ai/","language":"Python","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/AryanVBW.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":"CITATION.cff","codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null}},"created_at":"2023-12-08T01:48:21.000Z","updated_at":"2025-03-19T00:34:48.000Z","dependencies_parsed_at":"2023-12-18T04:30:39.665Z","dependency_job_id":null,"html_url":"https://github.com/AryanVBW/Private-Ai","commit_stats":null,"previous_names":["aryanvbw/private-ai"],"tags_count":4,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AryanVBW%2FPrivate-Ai","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AryanVBW%2FPrivate-Ai/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AryanVBW%2FPrivate-Ai/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AryanVBW%2FPrivate-Ai/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/AryanVBW","download_url":"https://codeload.github.com/AryanVBW/Private-Ai/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247698111,"owners_count":20981299,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","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":["aryanvbw","chatgpt","chatgpt-4","chatgpt3","chatgpt4","gpt4","gpt4all","privategpt","vivek","vivek-w"],"created_at":"2024-11-06T05:20:38.854Z","updated_at":"2025-04-07T17:33:33.179Z","avatar_url":"https://github.com/AryanVBW.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\n   \u003ca href=\"https://github.com/AryanVBw\"\u003e\n\u003cimg src=\"https://github.com/AryanVBW/Private-Ai/releases/download/I1/Bgdark.png\" alt=\"Darkside\"\u003e\u003c/a\u003e\u003c/p\u003e\n\n# 🚀 Welcome to Private-AI!\n\nPrivate-AI is an innovative AI project designed for asking questions about your documents using powerful Large Language Models (LLMs). The unique feature? It works offline, ensuring 100% privacy with no data leaving your environment.\n\n## 🌐 What does Private-AI offer?\n\n- **High-level API:** Abstracts the complexity of a Retrieval Augmented Generation (RAG) pipeline. Handles document ingestion, chat, and completions.\n  \n- **Low-level API:** For advanced users to implement custom pipelines. Includes features like embeddings generation and contextual chunks retrieval.\n## 🌟 Why Private-AI?\n\nPrivacy is the key motivator! Private-AI addresses concerns in data-sensitive domains like healthcare and legal, ensuring your data stays under your control.\n\n# 🤖 installation \n\n---\n\n**Private-Ai Installation Guide**\n\n- Install Python 3.11 (or 3.12)\n- Using apt(Debian base linux like-kali,Ubantu etc. )\n  ```bash\n  sudo apt-get install python3.11\n  apt install python3.11-venv\n  \n  ```\n- Using pyenv:\n  ```bash\n  pyenv install 3.11\n  pyenv local 3.11\n  ```\n\n- Install [Poetry](https://python-poetry.org/docs/#installing-with-pipx) for dependency management.\n```bash\n   sudo apt install python3-poetry\n  sudo apt install python3-pytest\n  ```\n### Installation Whithout GPU:\n- Git clone Private-Ai repository:\n```bash\ngit clone https://github.com/AryanVBW/Private-Ai\ncd Private-Ai \u0026\u0026 \\\npython3.11 -m venv .venv \u0026\u0026 source .venv/bin/activate \u0026\u0026 \\\npip install --upgrade pip poetry \u0026\u0026 poetry install --with ui,local \u0026\u0026 ./scripts/setup \npython3.11 -m private_gpt\n```\n\n### Run of private Ai:\n - forRunAgain jutsGoTo Private Ai   directoy anr run following comand:\n\n```bash\n  make run\n```\n# 👍👍All Done 👍👍\n\n## For GPU utilization and customization, follow the steps below:\n  - For Private-Ai to run fully locally GPU acceleration is required (CPU execution is possible, but very slow)\n### clone repo\n- Git clone Private-Ai repository:\n```bash\ngit clone https://github.com/AryanVBW/Private-Ai\ncd Private-Ai\n```\n### Dependencies Installation:\n- Install make (OSX: `brew install make`, Windows: `choco install make`).\n- Install dependencies:\n  ```bash\n  poetry install --with ui\n  ```\n### Local LLM Setup:\n- Install extra dependencies for local execution:\n  ```bash\n  poetry install --with local\n  ```\n- Use the setup script to download embedding and LLM models:\n  ```bash\n  poetry run python scripts/setup\n  ```\n### Finalize:\n- Installation of private Ai:\n  ```bash\n  make \n  ```  \n### Verification and run :\n- Run `make run` or `poetry run python -m private_gpt`.\n- Open http://localhost:8001 to see Gradio UI with a mock LLM echoing input.\n\n### Customization:\n- Customize low-level parameters in `private_gpt/components/llm/llm_component.py`.\n- Configure LLM options in `settings.yaml`.\n\n### GPU Support:\n- **OSX**: Build llama.cpp with Metal support.\n  ```bash\n  CMAKE_ARGS=\"-DLLAMA_METAL=on\" pip install --force-reinstall --no-cache-dir llama-cpp-python\n  ```\n\n- **Windows NVIDIA GPU**: Install [VS2022](https://visualstudio.microsoft.com/vs/community/), [CUDA toolkit](https://developer.nvidia.com/cuda-downloads), and run:\n  ```powershell\n  $env:CMAKE_ARGS='-DLLAMA_CUBLAS=on'; poetry run pip install --force-reinstall --no-cache-dir llama-cpp-python\n  ```\n\n- **Linux NVIDIA GPU and Windows-WSL**: Install [CUDA toolkit](https://developer.nvidia.com/cuda-downloads) and run:\n  ```bash\n  CMAKE_ARGS='-DLLAMA_CUBLAS=on' poetry run pip install --force-reinstall --no-cache-dir llama-cpp-python\n  ```\n\n### Troubleshooting:\n- Check GPU support and dependencies for your platform.\n- For C++ compiler issues, follow troubleshooting steps.\n\n**Note**: If any issues, retry in verbose mode with `-vvv` during installations.\n\n**Troubleshooting C++ Compiler**:\n- **Windows 10/11**: Install Visual Studio 2022 and MinGW.\n- **OSX**: Ensure Xcode is installed or install clang/gcc with Homebrew.\n\n---\n\n\n## 🧩 Architecture Highlights:\n\n- **FastAPI-Based API:** Follows the OpenAI API standard, making it easy to integrate.\n  \n- **LlamaIndex Integration:** Leverages LlamaIndex for the RAG pipeline, providing flexibility and extensibility.\n\n\n\n\n- **Present and Future:** Evolving into a gateway for generative AI models and primitives. Stay tuned for exciting new features!\n\n## 💡 How to Contribute?\n\nContributions are welcome! Check the ProjectBoard for ideas. Ensure code quality with format and typing checks (run `make check`).\n\n## 🤗Supporters:\n\nSupported by Qdrant, Fern, and LlamaIndex. Influenced by projects like LangChain, GPT4All, LlamaCpp, Chroma, and SentenceTransformers.\n\n👏 Thank you for contributing to the future of private and powerful AI with Private-AI!\n📝 **License:** Apache-2.0\n# Copyright Notice\nThis is a modified version of [PrivateGPT](https://github.com/imartinez/privateGPT). All rights and licenses belong to the PrivateGPT team.\n\n© 2023 PrivateGPT Developers. All rights reserved.\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faryanvbw%2Fprivate-ai","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Faryanvbw%2Fprivate-ai","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faryanvbw%2Fprivate-ai/lists"}