
by Omnara
Omnara is an open-source managed agent platform with one API for deploying and running AI agents on any model and any machine. You can self-host it under Apache 2.0 or use the hosted cloud.
APIContact for pricing
Step 1: Sign in at the Omnara cloud app, or self-host with Docker Compose.
Step 2: Write an agent config file that names the instruction, model and machine source.
Step 3: Launch it with npx omnara agents launch, or create it from the TypeScript SDK or REST API.
Step 4: Chat with the agent in the terminal or open it in the dashboard to watch tool calls.
Step 5: Connect Slack, Discord or GitHub, and set tool permissions so risky actions need approval.
The platform is free to use and self-host, with no platform fee. You pay only if you use Omnara-managed models or machines, which are billed as you go.
You can bring keys for providers such as OpenAI, Anthropic, OpenRouter and Amazon Bedrock, or point to endpoints you host yourself.
Yes. It is open source under Apache 2.0 and runs with Docker Compose. The README notes the local defaults are insecure and points to a self-hosting guide for real deployments.
On machines running the Omnara daemon, such as a laptop, VM or container, or on sandboxes from supported providers.
OlasOlas is a platform for owning and monetizing AI agents. It combines Pearl, a desktop app for running agents, with the Mech Marketplace, where agents offer and hire services.
DaytonaDaytona is an AI-focused api for developer tools workflows, giving users practical model-powered capabilities through a dedicated product or developer interface.