--- title: 'Kamal vs Blossom: A Detailed Comparison' url: "/docs/comparisons/kamal-vs-blossom/" canonical_url: https://docs.blossom88.com/docs/comparisons/kamal-vs-blossom/ markdown_url: https://docs.blossom88.com/docs/comparisons/kamal-vs-blossom.md summary: Compare Kamal and Blossom to find the best platform for your application deployment categories: - comparisons - platforms - kamal - deployment --- # Kamal vs Blossom: A Detailed Comparison There are significant differences between Kamal and Blossom. Kamal is the successor to Capistrano and is developed by the Rails team. Essentially, Kamal is "Capistrano with Containers." Blossom is a managed App platform that owns the ongoing deployment, health, scaling, and recovery workflow. ## Kamal ### How Kamal Works Kamal works by SSHing into multiple servers to install Docker and deploy your app. You configure Kamal settings, such as servers, using a `config/deploy.yml` file. It typically looks something like this: ```yaml service: app image: user/app servers: web: - 1.1.1.1 proxy: ssl: true host: app.example.com registry: username: user password: - KAMAL_REGISTRY_PASSWORD env: secret: - RAILS_MASTER_KEY - DATABASE_URL ``` ### Build and Deployment Process When you run `kamal deploy`, Kamal builds a container image using your local Dockerfile. This means your machine's architecture must match your deployment servers — otherwise, the image won't run. You can work around this by using a remote builder (`build.remote = true`) or running Kamal in CI/CD instead of locally. But both approaches add complexity like Docker layer caching not being used. It's also one more thing to manage. ### Managed Capacity Kamal does not provision or create servers for you, as this falls outside its scope. Similarly, it doesn't manage load balancers. You're also responsible for configuring this yourself. Kamal recommends using cloud provider managed load balancers and adding your servers manually. ### Container Management and Limitations Kamal uses vanilla `docker run` commands to start your containers. The specific options used are only visible during `kamal deploy`. This can make debugging challenging, as you might not know the exact options applied. While you can use `docker inspect` to reverse-engineer the options, it adds an extra step to the troubleshooting process. Kamal's support for multiple applications can feel like an afterthought. This might be due to the nature of having a `config/deploy.yml` in different repositories. This often leads to duplication in the `deploy.yml`, especially for configurations like `servers`. ### Overall Assessment To be clear, Kamal is a valuable tool for what it does: automating container image building and deployment to multiple servers. It allows deployment to any server accessible via SSH, offering substantial infrastructure flexibility. For many, its most significant advantage is that it's free. ### Key Features Summary Here's a summary of Kamal's key features: - It's "Capistrano with Containers" - It builds a container image on the deployment machine - It deploys your app to multiple servers via SSH - Does not manage load balancers - Requires manual server provisioning and maintenance - Configuration via YAML files for deployment targets - Only supports Dockerfile-based builds - Requires more DevOps knowledge and hands-on server management - Free to use ## Blossom Blossom extends beyond a deployment command into an ongoing App lifecycle. ### Product Reliability Kamal is command-driven: it connects to servers and runs deployment steps. Blossom records the intended Product state, evaluates observed health, and safely brings the App back toward that outcome. That is why Deploys, recovery, and scaling share one health model and durable Activity history. The customer follows the Product outcome rather than a sequence of host commands. ### App And Capacity Customers operate Apps, Environments, Services, Jobs, managed Data, and [Capacity](/docs/capacity/). [Blossom Cloud includes cloud costs; Blossom BYOC (bring your own cloud) uses a connected provider account and direct provider billing](/docs/locations/cloud-providers/). The App workflow stays the same. ### Deployment Approach Blossom records each desired Release. Deploy logs explain progress, health shows whether the Release is ready, and Activity preserves what changed. ### Build System and Container Technology A Dockerfile is optional. Pack, the static-site builder, an explicit Dockerfile, and pre-built images cover different source contracts. Customers can inspect the resulting Containers without managing their hosts. ### Pricing Model Blossom is a paid service. It publishes standard Product and Capacity rates, prorates charges by active time, and keeps paid Capacity within one capacity limit. ### Key Features Summary Here's a summary of Blossom's key features: - **Language and provider agnostic** — Deploy supported application stacks through one Product - **Fully managed PaaS** — UI and git-based workflows with automated deployments - **Start with your App** — No servers to set up first - **Flexible Build system** — Supports Pack, Dockerfile, static sites, or pre-built images - **Cloud Provider support** — Choose Blossom Cloud or Blossom BYOC - **Built-in operations** — Deploys, Container scaling, shell, logs, metrics, Activity, and Git provider integration - **Multiple Apps** — Keep each App's Environments, health, Activity, usage, and cost distinct - **Reliability platform** — Health-aware Deploys, recovery, scaling, and diagnostics - **Affordable pricing** — Commercial product with published standard pricing prorated by active time ## Quick Comparison | Feature | Kamal | Blossom | | ------------------- | --------------------- | ---------------------------- | | Deployment | CLI and YAML-based | **UI and git-based** | | Operating Responsibility | Customer manages deployment targets | **Blossom operates the platform** | | Capacity | Hosts provisioned and managed separately | **Regional Capacity policy managed by Blossom** | | Public Routing | Manual configuration | **Derived from App settings** | | Build Process | Dockerfile only | **Multiple build options** | | Operating Model | SSH command execution | **Managed Product lifecycle** | | Recovery | Rerun deployment or customer automation | **Blossom restores toward declared Product health** | | Scaling | Update targets and deployment configuration | **Service scaling and managed Capacity** | | Automation | Run Kamal directly or from CI | **UI and git-triggered Deploys** | | Setup Path | Prepare hosts, registry, and deployment configuration | **Create an App, choose a region, and Deploy** | | Pricing | Open-source tool plus infrastructure | **Published Product and Capacity rates** | ## Conclusion As evident, Kamal and Blossom differ significantly. It's really apples and oranges. People will inevitably compare the two. **Kamal** is a command-line deployment tool and the spiritual successor to Capistrano. Developed by the Rails team, it is designed for developers who want to deploy Containers while retaining responsibility for their servers. It is free and open source. **Blossom** is a managed App platform for teams that want to spend less time operating deployment infrastructure. Teams declare Product outcomes and use health plus Activity to verify them while Blossom handles the ongoing delivery lifecycle.