All your work.
Entirely on your terms.

Yes. Copy the full setup prompt in the deployment section into an assistant with terminal access. It asks whether to use AWS, GCP, an existing server or local evaluation, checks access without asking for secrets in chat, and asks for approval before creating paid resources or changing DNS. The prompt includes installation, HTTPS verification and a backup and upgrade handoff. Review what your assistant proposes and verify the result; you remain responsible for your infrastructure and its costs.

No. On a Linux host with root or sudo, the installer offers to install Docker Engine and the Compose plugin if they are missing. The README requires at least 2 GB RAM and 25 GB disk; 2 vCPU and 4 GB RAM are a small-team starting point. For the bundled HTTPS proxy, ports 80 and 443 must be free and your domain’s DNS A record must already point to the server. Leave the domain blank only for plain-HTTP evaluation.

Every self-hosted plan includes unlimited manual tickets/tasks, projects, QA test cases, test suites and plans, test runs, and documents and pages. The free core provides one workspace with Leera branding, without a paid license or registration for manual work. AI needs a configured provider; creating tickets/tasks, test cases, or projects through AI also requires a free registration and follows the instance license allowances. Your server, storage, email and provider usage are your operating costs. White-label adds custom branding for that one workspace; any broader Enterprise scope must be agreed separately.

The free self-hosted core has no per-seat software subscription. The standard White-label offer is licensed per installation; adding a teammate does not add a cloud seat subscription. Your infrastructure still needs enough capacity for the people and work it serves. Check the current pricing page and the terms shown in Instance Settings → License for the offer that applies to your installation.

Yes. Use the Docker Compose installer on a Linux server or the Helm chart in an existing Kubernetes cluster. Both deploy the application with its supporting services. The Compose installer can help install Docker and configures automatic HTTPS when your domain and network are ready. For production, plan persistent storage, service access, backups and capacity. The current API runs as one replica for consistent realtime updates; the web tier can scale separately.

Yes. Instance Settings → AI lets you configure providers, model names and endpoints, then test the connection. Supported provider types include OpenAI-compatible services, Anthropic, Google Gemini and Vertex AI. A compatible local endpoint, such as Ollama or vLLM, can run on your network. The selected models must support the workflows you want to use and be reachable from Leera. You cover provider usage or the infrastructure running your local models. Core workspace features remain available without AI configured.

The self-hosted core can run without registration, and paid offline licenses can be arranged for environments without outbound licensing connections. A connected deployment may contact licensing and update services, fetches the posts Leera publishes to the instance’s public blog, and reads the desktop app release list from GitHub for the sidebar’s Desktop app entry (server releases after 1.3.0). AI, email, login providers and integrations also need reachable endpoints, which can be internal where a compatible service is available. Plan how installation images and updates will reach an isolated environment, and discuss the network requirements with us before rollout.

Cloud and standard Self Hosted use Leera branding, including the Leera AI name. White-label Self Hosted removes the Leera name and vendor links and lets you configure the product name, brand color, application logo and favicon. The one exception is the desktop app download, once the desktop apps are released. On server releases after 1.3.0, the sidebar of a self-hosted instance then offers Leera Self-Hosted, which keeps its name, from Leera’s GitHub releases. A desktop app with a White-label or Enterprise instance’s own name and icon is planned and not yet available. Standard self-hosting and White-label each include one workspace per installation. Discuss any broader Enterprise scope with us before purchasing. The AI lotus icon stays the same across every plan, including White-label and Enterprise. Configure your identity in Instance Settings → Branding and review your entitlements in Instance Settings → License. Your public domain, email provider and storage configuration are part of the deployment setup; they do not depend on purchasing branding.

Yes. The default stack bundles PostgreSQL and SeaweedFS, an S3-compatible object store, and the deployment guide covers using external PostgreSQL and S3-compatible object storage. Configure the database connection for your deployment and set the storage details in Instance Settings. External file links need a reachable, separate storage origin without application credentials and response protections for untrusted uploads. The current bundled proxies apply those response protections to their own storage routes. Verify signed downloads and supported previews, and plan data transfer and backups before changing services on an instance that already contains work.

Your operating team owns the recovery plan, off-server backup copies, restore testing and update schedule. The Compose backup command captures the database, matching secret key and deployment configuration, and attempts to copy bundled object storage; verify that uploaded files are included, and protect external storage separately. With the updater enabled, Instance Settings → Updates can back up, download, migrate, restart and check the application. Server commands and Helm upgrades are also documented. Returning to older images does not undo database migrations, so keep a tested backup.

Lifetime marks a perpetual license with no license end date. Update coverage is listed separately and can have a fixed window. Releases published within that window remain eligible, and an installed version keeps running after the update window ends. Newer releases may require renewed update coverage. Check the current offer on the pricing page and the exact dates in Instance Settings → License; hosting and AI usage are separate from the license.

Your existing data, sign-in and workspaces remain accessible. Custom branding requires an active entitlement. If a verified White-label or Enterprise license expires, the app uses a neutral Workspace identity and continues hiding the Leera name and vendor links, apart from the sidebar’s desktop app download (server releases after 1.3.0), which names Leera Self-Hosted on every self-hosted plan once the desktop apps are released. Creating additional workspaces requires an eligible license. License checks do not lock your team out of workspaces that already exist. Update eligibility follows the update window on your license, which is separate from access to your existing work.

Yes: Leera Self-Hosted, for macOS 12 or later, Windows, Linux. Get it from the Desktop app entry at the bottom of your instance’s sidebar in a desktop browser (self-hosted server releases after 1.3.0 include it), or from the desktop releases of leera-app/leera-selfhost on GitHub. When it first opens, people enter your instance’s address—the same HTTPS URL your team opens in the browser—and notifications and other native features are granted only to that server. People sign in with an emailed code or their password. Each computer downloads the app and its updates from GitHub releases (github.com, which redirects to release-assets.githubusercontent.com) and checks leera.io for updates, so allow all three hosts on managed networks. The app keeps the Leera Self-Hosted name on every plan.