base on Container runtimes on macOS (and Linux) with minimal setup  ## Colima - container runtimes on macOS (and Linux) with minimal setup. [](https://github.com/abiosoft/colima/actions/workflows/go.yml) [](https://github.com/abiosoft/colima/actions/workflows/integration.yml) [](https://goreportcard.com/report/github.com/abiosoft/colima)  **Website & Documentation:** [colima.run](https://colima.run) | [colima.run/docs](https://colima.run/docs/) ## Features Support for Intel and Apple Silicon macOS, and Linux - Simple CLI interface with sensible defaults - Automatic Port Forwarding - Volume mounts - Multiple instances - Support for multiple container runtimes - [Docker](https://docker.com) (with optional Kubernetes) - [Containerd](https://containerd.io) (with optional Kubernetes) - [Incus](https://linuxcontainers.org/incus) (containers and virtual machines) - GPU accelerated containers for AI workloads ## Getting Started ### Installation Colima is available on Homebrew, MacPorts, Nix and [Mise](http://github.com/jdx/mise). Check [here](docs/INSTALL.md) for other installation options. ```sh # Homebrew brew install colima # MacPorts sudo port install colima # Nix nix-env -iA nixpkgs.colima # Mise mise use -g colima@latest ``` Or stay on the bleeding edge (only Homebrew) ``` brew install --HEAD colima ``` ## Usage Start Colima with defaults ``` colima start ``` For more usage options ``` colima --help colima start --help ``` Or use a config file ``` colima start --edit ``` ## Runtimes On initial startup, Colima initiates with a user specified runtime that defaults to Docker. ### Docker Docker client is required for Docker runtime. Installable with `brew install docker`. ``` colima start docker run hello-world docker ps ``` You can use the `docker` client on macOS after `colima start` with no additional setup. ### Containerd `colima start --runtime containerd` starts and setup Containerd. You can use `colima nerdctl` to interact with Containerd using [nerdctl](https://github.com/containerd/nerdctl). ``` colima start --runtime containerd nerdctl run hello-world nerdctl ps ``` It is recommended to run `colima nerdctl install` to install `nerdctl` alias script in $PATH. ### Kubernetes kubectl is required for Kubernetes. Installable with `brew install kubectl`. To enable Kubernetes, start Colima with `--kubernetes` flag. ``` colima start --kubernetes kubectl run caddy --image=caddy kubectl get pods ``` #### Interacting with Image Registry For Docker runtime, images built or pulled with Docker are accessible to Kubernetes. For Containerd runtime, images built or pulled in the `k8s.io` namespace are accessible to Kubernetes. ### Incus <small>**Requires v0.7.0**</small> Incus client is required for Incus runtime. Installable with brew `brew install incus`. `colima start --runtime incus` starts and setup Incus. ``` colima start --runtime incus incus launch images:alpine/edge incus list ``` You can use the `incus` client on macOS after `colima start` with no additional setup. **Note:** Running virtual machines on Incus is only supported on m3 or newer Apple Silicon devices. ### AI Models (GPU Accelerated) <small>**Requires v0.10.0, Apple Silicon and macOS 13+**</small> Colima supports GPU accelerated containers for AI workloads using the `krunkit` VM type. **Note:** To use krunkit with colima, ensure it is installed. ``` brew tap slp/krunkit brew install krunkit ``` Setup and use a model. ``` colima start --runtime docker --vm-type krunkit colima model run gemma3 ``` Colima supports two model runner backends: - **Docker Model Runner** (default) — supports [Docker AI Registry](https://hub.docker.com/u/ai) and [HuggingFace](https://huggingface.co). - **Ramalama** — supports [HuggingFace](https://huggingface.co) and [Ollama](https://ollama.com) registries. The default registry is the Docker AI Registry. Models can be run by name without a prefix: ```sh colima model run gemma3 colima model run llama3.2 # HuggingFace (Docker Model Runner) colima model run hf.co/microsoft/Phi-3-mini-4k-instruct-gguf # Ollama (requires ramalama runner) colima model run ollama://gemma3 --runner ramalama ``` See the [AI Workloads documentation](https://colima.run/docs/ai/) for more details. ### Customizing the VM The default VM created by Colima has 2 CPUs, 2GiB memory and 100GiB storage. The VM can be customized either by passing additional flags to `colima start`. e.g. `--cpu`, `--memory`, `--disk`, `--runtime`. Or by editing the config file with `colima start --edit`. **NOTE**: Disk size can be increased after the VM is created. #### Customization Examples - create VM with 1CPU, 2GiB memory and 10GiB storage. ``` colima start --cpu 1 --memory 2 --disk 10 ``` - modify an existing VM to 4CPUs and 8GiB memory. ``` colima stop colima start --cpu 4 --memory 8 ``` - create VM with Rosetta 2 emulation. Requires v0.5.3 and macOS >= 13 (Ventura) on Apple Silicon. ``` colima start --vm-type=vz --vz-rosetta ``` ## Project Goal To provide container runtimes on macOS with minimal setup. ## What is with the name? Colima means Containers on [Lima](https://github.com/lima-vm/lima). Since Lima is aka Linux Machines. By transitivity, Colima can also mean Containers on Linux Machines. ## And the Logo? The logo was contributed by [Daniel Hodvogner](https://github.com/dhodvogner). Check [this issue](https://github.com/abiosoft/colima/issues/781) for more. ## Troubleshooting and FAQs Check [here](docs/FAQ.md) for Frequently Asked Questions, or visit the [online FAQ](https://colima.run/docs/faq/) for a searchable version. ## How to Contribute? Check [here](docs/CONTRIBUTE.md) for the instructions on contributing to the project. ## Community - [GitHub Discussions](https://github.com/abiosoft/colima/discussions) - [GitHub Issues](https://github.com/abiosoft/colima/issues) - [Announcements](https://colima.run/announcements/) - `#colima` channel in the CNCF Slack - New account: <https://slack.cncf.io/> - Login: <https://cloud-native.slack.com/> ## License MIT ## Sponsoring the Project If you (or your company) are benefiting from the project and would like to support the contributors, kindly sponsor. - [Github Sponsors](https://github.com/sponsors/abiosoft) - [Buy me a coffee](https://www.buymeacoffee.com/abiosoft) - [Patreon](https://patreon.com/colima) --- [<img src="https://uploads-ssl.webflow.com/5ac3c046c82724970fc60918/5c019d917bba312af7553b49_MacStadium-developerlogo.png" style="max-height: 150px"/>](https://macstadium.com) ", Assign "at most 3 tags" to the expected json: {"id":"2490","tags":[]} "only from the tags list I provide: []" returns me the "expected json"