Omnara: open-source managed agents
Omnara is an open-source API for running agents on any model and machine. Self-host under Apache 2.0, or use the cloud with a $0 platform fee.
Key features
- Runs agents through one REST API, a TypeScript SDK, and a CLI
- Uses OpenAI, Anthropic, OpenRouter, LiteLLM, or a model you host
- Runs agents on sandboxes or on your own macOS or Linux machine
- Stores each agent's steps in a durable event log so a crash can resume
- Requires approval for risky tool calls and keeps an audit history
- Connects Slack, Discord, GitHub, MCP servers, and skills
- Self-host with Docker Compose, or use Omnara Cloud
- Apache 2.0. The cloud platform fee is $0 with your own keys and machines
- Category
- Agents
- Who it's for
- Teams who want to run production agents on their own models and machines.
- Tags
- agentapiterminalllm-apitool-callingsandboxhostingcontainerserror-trackingjavascriptversion-controlself-hosted
- Website
- omnara.com
- Listed
- 2026-10-22
Frequently asked questions
Is Omnara free?
Omnara has a free plan, with paid tiers for more usage or features.
How much does Omnara cost?
Platform fee $0; pay as you go for models and machines, according to its pricing page.
What is Omnara used for?
Open-source agent API for any model and machine. Self-host free under Apache 2.0, or use the cloud with a $0 platform fee. Typical uses: runs agents through one REST API, a TypeScript SDK, and a CLI; uses OpenAI, Anthropic, OpenRouter, LiteLLM, or a model you host; runs agents on sandboxes or on your own macOS or Linux machine.
What are the best Omnara alternatives?
Alternatives in the same category include Agensh, OpenShell, Docker Agentic Platform, E2B, Alter.
Related in Agents
See all →- Agensh Self-organized multi-agent harness. Agents claim sub-tasks and merge progress without a central orchestrator. MIT.
- Alter Privileged access for AI agents. Vaulted credentials, per-call policy, and an audit log. Free tier is 10,000 API calls a month.
- Harness² Framework that refines an agent harness on new tasks while model weights stay fixed. Apache 2.0.
- LearnHarness Agent harness that plans and teaches from your current understanding, then keeps a knowledge graph. AGPL-3.0.
- OpenShell Safe, private runtime for fleets of autonomous AI agents. Sandboxes enforce file, syscall, and network policy. Apache 2.0.
- Agentic Batch Changes Describe a change in plain language. The agent scopes it across your indexed codebase and tracks it through to merge.