{"data":{"slug":"containers-ramalama","name":"ramalama","tagline":"Simplifies local serving of AI models through containers","github_url":"https://github.com/containers/ramalama","owner":"containers","repo":"ramalama","owner_avatar_url":"https://avatars.githubusercontent.com/u/5874934?v=4","primary_language":"Python","stars":3053,"forks":367,"topics":["ai","containers","cuda","hacktoberfest","hip","inference-server","intel","llamacpp","llm","podman","vllm"],"archived":false,"github_pushed_at":"2026-09-19T00:03:50+00:00","maintenance_label":"Very active","stars_delta_30d":53,"url":"https://www.graphcanon.com/tools/containers-ramalama","markdown_url":"https://www.graphcanon.com/tools/containers-ramalama.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/containers-ramalama","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=containers-ramalama","description":"RamaLama is an open-source developer tool that simplifies the local serving of AI models from any source and facilitates their use for inference in production, all through the familiar language of containers.","homepage_url":"https://ramalama.ai","license":"MIT","open_issues":115,"watchers":33,"ai_summary":"RamaLama is an open-source tool that simplifies the process of setting up and running AI model inference with container technology.","readme_excerpt":"### Install on macOS (Self-Contained Installer)\nDownload the self-contained macOS installer that includes Python and all dependencies:\n\n1. Download the latest `.pkg` installer from [Releases](https://github.com/containers/ramalama/releases)\n2. Double-click to install, or run: `sudo installer -pkg RamaLama-*-macOS-Installer.pkg -target /`\n\nSee [macOS Installation Guide](docs/MACOS_INSTALL.md) for detailed instructions.\n\n---\n\n### Install on Fedora\nRamaLama is available in [Fedora](https://fedoraproject.org/) and later. To install it, run:\n```\nsudo dnf install ramalama\n```\n\n---\n\n### Install via PyPI\nRamaLama is available via PyPI at [https://pypi.org/project/ramalama](https://pypi.org/project/ramalama)\n```\npip install ramalama\n```\n\n---\n\n### Install script (Linux and macOS)\nInstall RamaLama by running:\n```\ncurl -fsSL https://ramalama.ai/install.sh | bash\n```\n\n---\n\n### Install on Windows\nRamaLama supports Windows with Docker Desktop or Podman Desktop:\n```powershell\npip install ramalama\n```\n\n**Requirements:**\n- Python 3.9 or later\n- Docker Desktop or Podman Desktop with WSL2 backend\n- For GPU support, see [NVIDIA GPU Setup for WSL2](docs/readme/wsl2-docker-cuda.md)\n\n**Note:** Windows support requires running containers via Docker/Podman. The model store uses hardlinks (no admin required) or falls back to file copies if hardlinks are unavailable.\n\n---\n\n## Hardware Support\n\n| Hardware                           | Enabled                     |\n| :--------------------------------- | :-------------------------: |\n| CPU                                | ✓                     |\n| Apple Silicon GPU (Linux / Asahi)  | ✓                     |\n| Apple Silicon GPU (macOS)          | ✓ llama.cpp or MLX    |\n| Apple Silicon GPU (podman-machine) | ✓                     |\n| Nvidia GPU (cuda)                  | ✓ See note below      |\n| AMD GPU (rocm, vulkan)             | ✓                     |\n| Ascend NPU (Linux)                 | ✓                     |\n| Intel ARC GPUs (Linux)             | ✓ See note below      |\n| Intel GPUs (vulkan / Linux)        | ✓                     |\n| Moore Threads GPU (musa / Linux)   | ✓ See note below      |\n| Windows (with Docker/Podman)       | ✓ Requires WSL2       |","github_created_at":"2024-07-24T19:09:58+00:00","created_at":"2026-07-15T11:19:12.806449+00:00","updated_at":"2026-09-20T05:19:19.241196+00:00","categories":[{"slug":"developer-tools","name":"Developer Tools","url":"https://www.graphcanon.com/categories/developer-tools","markdown_url":"https://www.graphcanon.com/categories/developer-tools.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/developer-tools"},{"slug":"inference-serving","name":"Inference & Serving","url":"https://www.graphcanon.com/categories/inference-serving","markdown_url":"https://www.graphcanon.com/categories/inference-serving.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/inference-serving"}],"tags":[{"slug":"ai","name":"ai"},{"slug":"containers","name":"containers"},{"slug":"cuda","name":"cuda"},{"slug":"hip","name":"hip"},{"slug":"inference-server","name":"inference-server"},{"slug":"intel","name":"intel"},{"slug":"llamacpp","name":"llamacpp"},{"slug":"podman","name":"podman"}],"trust":{"provenance":{"is_fork":false,"github_id":833306239,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-20T05:19:16.609Z","maintenance":{"label":"Very active","score":96,"methodology":"github_public_v1","releases_90d":2,"days_since_push":1,"last_release_at":"2026-08-21T15:29:44Z","stars_delta_30d":53,"open_issues_delta_30d":7},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-15T11:19:14.176Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-09-20T05:19:17.860Z"},"has_cli":{"value":true,"source":"pyproject.toml:[project.scripts]","observed_at":"2026-09-20T05:19:17.860Z"},"languages":{"value":["python"],"source":"github.language+pyproject.toml","observed_at":"2026-09-20T05:19:17.860Z"},"license_spdx":{"value":"MIT","source":"github.license","observed_at":"2026-09-20T05:19:17.860Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["When you need to serve multiple AI models locally across various accelerators like CPUs, GPUs (Apple Silicon, Nvidia, AMD), Arc GPUs, Ascend NPU, and Moore Threads for rapid inference.","For macOS users requiring a self-contained installer that bundles Python and essential dependencies into one executable."],"when_not_to_use":["Avoid using RamaLama if you prefer native OS integration over containerization, as it relies heavily on Docker or Podman technology.","If your project strictly avoids the MIT license for compliance reasons, look elsewhere since all of RamaLama's flexibility comes under this licensing scheme."],"source":"enrich:decision_facts","observed_at":"2026-07-17T11:36:15.707Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"RamaLama simplifies local AI model serving through containers with versatile hardware support, including Apple Silicon, Nvidia CUDA, AMD ROCm, Intel ARC GPUs, Ascend NPU, and Moore Threads MUSA."}]}}