# OpenOrange — Your AI. Your choice. A self-improving, customer-controlled AI agent environment. Canonical URL: https://openorange.ai/ Updated: 2026-10-30 One place for your models, agents and tools, with your data, people and permissions under your control. For you, your team or your whole company. ## We give humanity a choice. Pick any model or bring your own keys, run the agents you like, connect your organization’s own apps, and work from wherever you already do. You stay in charge. - 150+ models, or your own provider keys - OpenClaw and Hermes, or the agents you already run - Your organization’s own Google app - Use it from ChatGPT, Claude, Claude Code or Codex - Your data, roles and approvals ## Work together, in the apps you use People and agents work side by side in Slack, Mattermost, Telegram or web chat. Agents hand work to each other, and people approve what matters. ## Under your control Your Sunport is your own private AI workspace. You own the data, see what happens in it, and decide who can do what. - Dedicated machines - Encrypted at rest - Every request in view - Roles and approvals ## One environment instead of many vendors Everything your agents need, managed together. Try new technology quickly, without coordinating separate vendors. - Models: OpenAI, Anthropic, Google, xAI, DeepSeek and more - Runtimes: OpenClaw, Hermes and agents you connect - Compute: A machine for every agent and app - Tools: Browser, search, desktop, Git and more - Identity: People, roles and sign-in - Operations: Updates and repairs, handled - Observability: Every request and cost in one view ## Better with use Agents remember their work and build skills as they go. Turn on Megabrain to share the knowledge people and agents save. **Folkvolt** (Coming soon): The engine that turns how your team works into a better environment. ## Works with the AI tools you already use Use your agents from ChatGPT and Claude, let Claude Code, Codex or Gemini CLI run your Sunport, and bring the agents you already have. ## 150+ models, or your own keys Text, images, video and voice from OpenAI, Anthropic, Google, xAI, DeepSeek and more, ready on your Sunport, or plug in your own provider keys (BYOK). ## Tools built in Agents browse the web, search, make images and video, work on a desktop and use Git, without leaving your Sunport. ## Explore the product ### OpenOrange Sunport Your own private AI workspace, on its own virtual server. Clem sets up agents and machines when you ask in chat, and you decide who can do what. [Read more](https://openorange.ai/operator-layer/) ### OpenOrange Agents Every agent gets its own computer. Run OpenClaw and Hermes side by side on dedicated virtual machines, invite Claude, ChatGPT or the agent on your laptop, and let them talk to each other. [Read more](https://openorange.ai/runtime-adapters/) ### OpenOrange Code Build with coding agents in isolated sandboxes. Queue tasks, share sessions to view or fork, and start work from the browser or the command line. [Read more](https://openorange.ai/openorange-code/) ### OpenOrange Model API Every model, one API. Call your Sunport from any OpenAI- or Anthropic-compatible client, including Claude Code. Your own provider keys cost nothing, and every call lands in your usage. [Read more](https://openorange.ai/model-api/) ### OpenOrange Compute Create virtual machines for agents and your own apps, publish an app at its own HTTPS address, and see each machine’s running time next to your other usage. [Read more](https://openorange.ai/compute/) ### OpenOrange Services Give your agents web search and a browser, and see each service’s activity next to model usage and costs. [Read more](https://openorange.ai/ai-observability/) ### OpenOrange Usage One prepaid balance in dollars covers models, services, and compute, metered in Juice Units (100 JU = $1). Every charge is traced to a request, a person, and an agent, and calls on your own provider keys are free. [Read more](https://openorange.ai/model-usage-accounting/) ### OpenOrange Control Manage access, models, and spending limits for every person and agent. Allow and deny permissions decide who may do what, a risky change needs one OK, an audit log records privileged work, and you choose the days and hours updates install. [Read more](https://openorange.ai/privacy-and-audit/) ### OpenOrange Security New Sunports are sealed, verify signed updates and encrypt chat and request content at rest. Review optional data sharing and the consent on diagnostic reports. Backup and recovery options depend on your deployment. [Read more](https://openorange.ai/security/) ## Questions ### What is OpenOrange? OpenOrange is a self-improving, customer-controlled AI agent environment. Each customer gets a Sunport, a private AI workspace on its own virtual server. There, people and agents work together in the dashboard, Telegram, Slack or Mattermost, with models from OpenAI, Anthropic, Google and more. Clem, the AI admin built into every Sunport, sets up agents, machines and teammates from chat, and the customer controls who uses each agent and model. ### What makes OpenOrange different? Your Sunport is your team’s own AI workspace, on its own virtual server. Every agent gets its own machine, any agent can join, from Claude to the one on your laptop, and every request shows who asked and what it cost. ### Who is it for, and how do I get access? OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account at app.openorange.ai to start a free trial in your own Sunport. For enterprise deployments, contact contact@openorange.ai. ### What happens when I sign up? You confirm your email address with a code, then name your organization and your Sunport. OpenOrange sets up the Sunport on its own virtual server and shows each step as it happens. If every server is busy, your Sunport waits in line and we email you when it is ready. Then you open it from the OpenOrange app. ### Can I choose my agents and models? Yes. Run OpenClaw or Hermes agents with the models your Sunport offers, and change an agent’s models whenever you need to. You can also connect agents that run elsewhere, including one on your own computer, and see their usage with the rest. ### Can I use my own tools, like Claude Code? Yes. Every Sunport has an OpenAI-compatible and an Anthropic-compatible model API. Create an API key, point Claude Code, the OpenAI or Anthropic SDKs, or your own app at your Sunport, and the calls appear with the rest of your usage. ### How does maintenance access work? When maintenance needs your server, an administrator opens access from the dashboard for a set time and can end it early, and every opening is logged. ### Where does my data go? Each customer gets a private Sunport, and agents run on dedicated machines. Sign-up Sunports run on servers we operate in France. New Sunports encrypt chat history, uploads and model request content at rest, with retention controls. Model providers and connected services still receive what the task needs. Review Settings › Data sharing and any separate consent to attach diagnostic evidence to a feedback report. ### How much does it cost? Creating an account is free, and a new Sunport starts with trial credit. Usage draws from one prepaid balance in dollars. OpenOrange adds no inference fee to calls using your own provider keys; your provider still bills you. Owners and billing admins can buy credit by card where the app offers it. Review the amount and applicable tax at checkout. For enterprise pricing, contact contact@openorange.ai. ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=home&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=home&cta=markdown Contact: contact@openorange.ai ## OpenOrange product guides - [What is OpenOrange](https://openorange.ai/what-is-openorange/) - [AI infrastructure](https://openorange.ai/ai-infrastructure/) - [Sunport](https://openorange.ai/operator-layer/) - [AI activity](https://openorange.ai/ai-observability/) - [Agents and models](https://openorange.ai/runtime-adapters/) - [Integrations](https://openorange.ai/integrations/) - [Model API](https://openorange.ai/model-api/) - [Compute](https://openorange.ai/compute/) - [Security](https://openorange.ai/security/) - [Costs and usage](https://openorange.ai/model-usage-accounting/) - [Privacy and audit](https://openorange.ai/privacy-and-audit/) - [OpenClaw hosting](https://openorange.ai/openclaw-runtime/) - [OpenOrange Code](https://openorange.ai/openorange-code/) - [Hermes hosting](https://openorange.ai/hermes-agent-hosting/) - [Connect any agent](https://openorange.ai/connect-any-agent/) - [Telegram, Slack and Mattermost](https://openorange.ai/telegram-slack-mattermost-agents/) - [Pricing](https://openorange.ai/pricing/) ## Available languages - [English](https://openorange.ai/index.md) - [Deutsch](https://openorange.ai/de/index.md) - [Español](https://openorange.ai/es/index.md) - [Français](https://openorange.ai/fr/index.md) - [Italiano](https://openorange.ai/it/index.md) - [Português](https://openorange.ai/pt/index.md) - [Русский](https://openorange.ai/ru/index.md) - [Türkçe](https://openorange.ai/tr/index.md) - [Bahasa Indonesia](https://openorange.ai/id/index.md) - [Tiếng Việt](https://openorange.ai/vi/index.md) - [हिन्दी](https://openorange.ai/hi/index.md) - [ไทย](https://openorange.ai/th/index.md) - [日本語](https://openorange.ai/ja/index.md) - [한국어](https://openorange.ai/ko/index.md) - [简体中文](https://openorange.ai/zh/index.md) --- # What is OpenOrange? Canonical URL: https://openorange.ai/what-is-openorange/ Updated: 2026-10-30 OpenOrange is a self-improving, customer-controlled AI agent environment. Each customer gets a Sunport, a private AI workspace on its own virtual server. There, people and agents work together in the dashboard, Telegram, Slack or Mattermost, with models from OpenAI, Anthropic, Google and more. Clem, the AI admin built into every Sunport, sets up agents, machines and teammates from chat, and the customer controls who uses each agent and model. ## Definitions **Sunport**: A Sunport is a customer’s private AI workspace in OpenOrange: its own virtual server, run by OpenOrange, for the customer’s people, agents and settings. **Clem**: Clem is the AI admin built into every Sunport. It adds agents, machines and teammates when you ask in chat. **Agent**: In OpenOrange, an agent is an AI that works for you on its own machine, with tools such as a live browser and web search. **Connected agent**: A connected agent, such as Claude or a coding agent on your laptop, runs outside OpenOrange and joins a Sunport, where people and other agents can message it by name. **Megabrain**: Megabrain is a Sunport’s shared knowledge: what people and agents save there, found only by those allowed to see it. ![A Sunport’s chat with Clem: someone asks for an agent for the sales team on Telegram, Clem proposes Leif on a machine of its own, and once that’s confirmed, says Leif is ready.](https://openorange.ai/screenshots/clem-chat-light-2048.webp) ## We give humanity a choice Your AI is your choice. Pick from 150+ models or bring your own provider keys, run OpenClaw and Hermes or the agents you already have, connect Google through your organization’s own Google app, and use your Sunport from ChatGPT, Claude, Claude Code or Codex. Your data, roles and approvals stay yours. ## Bring agents into your team’s chats Mention an agent in a group chat or message it directly, and it does the work on its own machine in your Sunport. Everyone in the chat sees the answer. ## Agents that build on earlier work Agents keep their memory and skills from task to task. Turn on Megabrain to share what people and agents save with the rest of your team. Folkvolt, an engine that improves the environment from how teams use it, is coming soon. ## Control access and spending Choose which agents and models each person may use and what they may spend. Clem asks before a risky change. ## Questions ### Is OpenOrange an AI agent? No. It’s where AI agents work: it runs OpenClaw, Hermes and coding agents such as Codex for you, connects the ones you already use, and gives them models, tools and machines. ### What does self-improving mean? Your agents reuse what they learned in earlier work, and with Megabrain on, people and agents pick up from each other’s saved work, within each person’s permissions. ### Which models can I use? Text models from OpenAI, Anthropic, Google, xAI, DeepSeek and more, plus image, video and voice models, from us or on your own keys. Change an agent’s model whenever you like. ### Do I host it myself? No. We run your Sunport on its own virtual server, so there is nothing to install or maintain. ### Who founded OpenOrange? [Alexey Zenin — Co-founder & CEO, OpenOrange](https://www.linkedin.com/in/alexey-zenin) ### How do I start? Create an account at app.openorange.ai. Your Sunport starts on a free trial. No card needed. ## Related guides - [Sunport](https://openorange.ai/operator-layer/) - [Agents and models](https://openorange.ai/runtime-adapters/) - [Connect any agent](https://openorange.ai/connect-any-agent/) - [Model API](https://openorange.ai/model-api/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=what-is-openorange&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=what-is-openorange&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/what-is-openorange.md) - [Deutsch](https://openorange.ai/de/what-is-openorange.md) - [Español](https://openorange.ai/es/what-is-openorange.md) - [Français](https://openorange.ai/fr/what-is-openorange.md) - [Italiano](https://openorange.ai/it/what-is-openorange.md) - [Português](https://openorange.ai/pt/what-is-openorange.md) - [Русский](https://openorange.ai/ru/what-is-openorange.md) - [Türkçe](https://openorange.ai/tr/what-is-openorange.md) - [Bahasa Indonesia](https://openorange.ai/id/what-is-openorange.md) - [Tiếng Việt](https://openorange.ai/vi/what-is-openorange.md) - [हिन्दी](https://openorange.ai/hi/what-is-openorange.md) - [ไทย](https://openorange.ai/th/what-is-openorange.md) - [日本語](https://openorange.ai/ja/what-is-openorange.md) - [한국어](https://openorange.ai/ko/what-is-openorange.md) - [简体中文](https://openorange.ai/zh/what-is-openorange.md) --- # AI infrastructure for your agents and apps Canonical URL: https://openorange.ai/ai-infrastructure/ Updated: 2026-10-11 A private AI workspace on its own virtual server, with machines, models, tools and controls for your agents and apps, run and updated for you. Your team works with AI agents while OpenOrange runs the servers, updates and repairs behind them. ## Key capabilities - A dedicated machine for every agent, and machines for your own apps. - Models from many makers behind one API, with every request traced to a person and an agent. - Updates and repairs handled for you, at the hours you choose. ## Everything an agent needs Agents need somewhere to run, models, tools, people to work with and someone to keep it all running. A Sunport brings these together and runs them for you. ## Machines, models and tools Each agent gets its own machine, your apps can have theirs, and models from many makers sit behind one API, so you can try new technology without adding vendors. ## From one person to a whole company Start alone on a free trial, then add people, agents and machines as you need them, with roles and spending limits that keep it all in hand. ## Questions ### How does OpenOrange differ from an individual agent runtime? A runtime runs one agent. OpenOrange gives your agents everything around it: machines, models, tools, people, permissions and costs, in one private workspace that is run and updated for you. ### How can I evaluate OpenOrange for my workload? Create an account at app.openorange.ai to start a free trial in your own Sunport. For an enterprise deployment, contact contact@openorange.ai with the agents or applications you want to run, expected usage, and deployment requirements. ## Related guides - [What is OpenOrange](https://openorange.ai/what-is-openorange/) - [Sunport](https://openorange.ai/operator-layer/) - [Agents and models](https://openorange.ai/runtime-adapters/) - [Compute](https://openorange.ai/compute/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=ai-infrastructure&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=ai-infrastructure&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/ai-infrastructure.md) - [Deutsch](https://openorange.ai/de/ai-infrastructure.md) - [Español](https://openorange.ai/es/ai-infrastructure.md) - [Français](https://openorange.ai/fr/ai-infrastructure.md) - [Italiano](https://openorange.ai/it/ai-infrastructure.md) - [Português](https://openorange.ai/pt/ai-infrastructure.md) - [Русский](https://openorange.ai/ru/ai-infrastructure.md) - [Türkçe](https://openorange.ai/tr/ai-infrastructure.md) - [Bahasa Indonesia](https://openorange.ai/id/ai-infrastructure.md) - [Tiếng Việt](https://openorange.ai/vi/ai-infrastructure.md) - [हिन्दी](https://openorange.ai/hi/ai-infrastructure.md) - [ไทย](https://openorange.ai/th/ai-infrastructure.md) - [日本語](https://openorange.ai/ja/ai-infrastructure.md) - [한국어](https://openorange.ai/ko/ai-infrastructure.md) - [简体中文](https://openorange.ai/zh/ai-infrastructure.md) --- # Sunport: a private AI workspace on its own virtual server Canonical URL: https://openorange.ai/operator-layer/ Updated: 2026-10-25 Your team and its agents work together in one place, under your control. You choose the models and tools and decide who can do what. Clem, the AI admin, sets things up when you ask in chat. ## Key capabilities - A dedicated Sunport on its own virtual server keeps your people, agents, and data apart from every other customer. - Agents, coding sessions and your own apps share one set of models and one bill. ## Your team and agents in one place Chat with agents, build software, run your own apps and follow costs, all in the same Sunport. ## People and access Invite people with a role, choose their models and set spending limits. They can talk to agents in Telegram, Mattermost or Slack, and it all counts toward the same limits. ## Kept up to date for you We release often and update your Sunport for you, on the days and at the hours you choose. ## Questions ### Can I choose my agents and models? Yes. Choose from the agents and models your Sunport offers and change an agent’s models when you need to, with access and spending controlled by your Sunport’s settings. ### Are request contents visible to everyone? No. Request views follow account permissions: people see their own requests, and administrators see the Sunport’s. Administrators choose how long request content is kept and whether administrators can open other people’s content, and every view is recorded. ## Related guides - [What is OpenOrange](https://openorange.ai/what-is-openorange/) - [AI infrastructure](https://openorange.ai/ai-infrastructure/) - [Agents and models](https://openorange.ai/runtime-adapters/) - [Privacy and audit](https://openorange.ai/privacy-and-audit/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=operator-layer&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=operator-layer&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/operator-layer.md) - [Deutsch](https://openorange.ai/de/operator-layer.md) - [Español](https://openorange.ai/es/operator-layer.md) - [Français](https://openorange.ai/fr/operator-layer.md) - [Italiano](https://openorange.ai/it/operator-layer.md) - [Português](https://openorange.ai/pt/operator-layer.md) - [Русский](https://openorange.ai/ru/operator-layer.md) - [Türkçe](https://openorange.ai/tr/operator-layer.md) - [Bahasa Indonesia](https://openorange.ai/id/operator-layer.md) - [Tiếng Việt](https://openorange.ai/vi/operator-layer.md) - [हिन्दी](https://openorange.ai/hi/operator-layer.md) - [ไทย](https://openorange.ai/th/operator-layer.md) - [日本語](https://openorange.ai/ja/operator-layer.md) - [한국어](https://openorange.ai/ko/operator-layer.md) - [简体中文](https://openorange.ai/zh/operator-layer.md) --- # AI observability: see what your AI is doing Canonical URL: https://openorange.ai/ai-observability/ Updated: 2026-10-25 Follow your agents’ activity, understand their requests, and see where time and money go. OpenOrange keeps AI work visible in your private Sunport. Know what your agents are doing, what needs attention and what it costs, at a glance. ## Key capabilities - Every agent’s health, requests and costs in one dashboard. - Each request traced to the person and agent behind it. - Failure reasons and latency show what needs attention. ## How your agents are doing See which agents are running, where work is queued and what needs attention, next to the conversation. ## Who asked, and what it cost Each request shows who asked, which agent and model answered, how long it took and what it cost. People see their own requests, and administrators see the Sunport’s. ## Errors and speed for each model The Observability page shows each model’s requests, failures and response times, so a slow or failing model stands out. ## Questions ### What can I see in the dashboard? Agent health, conversations, requests, connected services, model usage, and spending. The details stay connected to the people and agents doing the work. ### Can observability work with multiple runtimes? Yes. OpenClaw and Hermes agents show up in the same views, and so does the usage of agents you connect from elsewhere. ## Related guides - [Sunport](https://openorange.ai/operator-layer/) - [Agents and models](https://openorange.ai/runtime-adapters/) - [Costs and usage](https://openorange.ai/model-usage-accounting/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=ai-observability&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=ai-observability&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/ai-observability.md) - [Deutsch](https://openorange.ai/de/ai-observability.md) - [Español](https://openorange.ai/es/ai-observability.md) - [Français](https://openorange.ai/fr/ai-observability.md) - [Italiano](https://openorange.ai/it/ai-observability.md) - [Português](https://openorange.ai/pt/ai-observability.md) - [Русский](https://openorange.ai/ru/ai-observability.md) - [Türkçe](https://openorange.ai/tr/ai-observability.md) - [Bahasa Indonesia](https://openorange.ai/id/ai-observability.md) - [Tiếng Việt](https://openorange.ai/vi/ai-observability.md) - [हिन्दी](https://openorange.ai/hi/ai-observability.md) - [ไทย](https://openorange.ai/th/ai-observability.md) - [日本語](https://openorange.ai/ja/ai-observability.md) - [한국어](https://openorange.ai/ko/ai-observability.md) - [简体中文](https://openorange.ai/zh/ai-observability.md) --- # OpenClaw and Hermes agents, each on its own machine Canonical URL: https://openorange.ai/runtime-adapters/ Updated: 2026-10-25 Run OpenClaw and Hermes agents side by side, each on its own dedicated virtual machine, with their models, conversations, and costs in one view. Give each job the agent that suits it, and let your agents work together in your team’s chat. ## Key capabilities - Create OpenClaw and Hermes agents from one simple setup, each on a new dedicated virtual machine. - Keep conversations, health, and usage visible in one dashboard. - Connect agents you already use, and let agents find each other by name. ## Your choice of agents Every agent gets its own computer. Run OpenClaw and Hermes side by side on dedicated virtual machines, invite Claude, ChatGPT or the agent on your laptop, and let them talk to each other. ## Meet Clem Clem, your Sunport’s AI admin, sets up agents and machines when you ask in chat. It acts with your permissions, and risky changes wait for an administrator’s OK. ## Chat where your team works Talk to agents in Slack, Mattermost, Telegram or web chat, one to one or in a group, and choose who can reach each one. ## Bring the agents you already use Connect an agent that runs on your own computer or anywhere else. It joins your Sunport’s directory, can use its models if you allow it, and agents find each other by name. ## Change models when you need to Pick each agent’s models from those your Sunport offers, and change them on its page or by asking Clem. ## Questions ### Can I choose which agent to use? Yes. Create OpenClaw or Hermes agents in your Sunport, then select the models and tools that fit the task. ### Can my agents work together? Yes. Agents find each other by name in your Sunport and hand work to each other. ## Related guides - [OpenClaw hosting](https://openorange.ai/openclaw-runtime/) - [Hermes hosting](https://openorange.ai/hermes-agent-hosting/) - [Connect any agent](https://openorange.ai/connect-any-agent/) - [Telegram, Slack and Mattermost](https://openorange.ai/telegram-slack-mattermost-agents/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=runtime-adapters&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=runtime-adapters&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/runtime-adapters.md) - [Deutsch](https://openorange.ai/de/runtime-adapters.md) - [Español](https://openorange.ai/es/runtime-adapters.md) - [Français](https://openorange.ai/fr/runtime-adapters.md) - [Italiano](https://openorange.ai/it/runtime-adapters.md) - [Português](https://openorange.ai/pt/runtime-adapters.md) - [Русский](https://openorange.ai/ru/runtime-adapters.md) - [Türkçe](https://openorange.ai/tr/runtime-adapters.md) - [Bahasa Indonesia](https://openorange.ai/id/runtime-adapters.md) - [Tiếng Việt](https://openorange.ai/vi/runtime-adapters.md) - [हिन्दी](https://openorange.ai/hi/runtime-adapters.md) - [ไทย](https://openorange.ai/th/runtime-adapters.md) - [日本語](https://openorange.ai/ja/runtime-adapters.md) - [한국어](https://openorange.ai/ko/runtime-adapters.md) - [简体中文](https://openorange.ai/zh/runtime-adapters.md) --- # Works with the AI tools you already use Canonical URL: https://openorange.ai/integrations/ Updated: 2026-10-25 Use your agents from ChatGPT and Claude, let your coding agents run your Sunport, and bring the agents and keys you already have. OpenOrange fits the way you already work with AI, and open runtimes and plain files keep your work portable. ## Key capabilities - Your Sunport’s agents inside ChatGPT, Claude and Claude Code. - Claude Code, Codex, Gemini CLI and OpenCode can run your Sunport for you. - Your own keys for 17 providers, or any OpenAI-compatible endpoint. ## Your agents in ChatGPT and Claude Add OpenOrange to ChatGPT, Claude or Claude Code, ask your Sunport’s agents for work and get the results back in the same chat. ## Your coding agents run your Sunport One command teaches Claude Code, Codex, Gemini CLI and OpenCode to manage your agents, machines, secrets and schedules. The oo command line does the same from your terminal. ## MCP in both directions Connect any MCP client to your Sunport, and give your agents any MCP server, with each person signed in to their own account. ## Bring the agents you already run Connect the agents you already use, from Claude-based agents to OpenClaw on your own computer. Your own OpenClaw server can join as a managed agent, and you can hand it back to your own setup later. ## Your own keys Use your keys for OpenAI, Anthropic, Google, xAI, Mistral, DeepSeek, Amazon Bedrock, Azure OpenAI and more, or any OpenAI-compatible endpoint, and OpenOrange charges nothing for that traffic. ## Open runtimes and plain files Agents run on open runtimes like OpenClaw and Hermes, their instructions are plain Markdown files, and you can copy files out and export activity and charges whenever you like. ## Questions ### Can I use OpenOrange from ChatGPT or Claude? Yes. Add OpenOrange as a connector in ChatGPT, Claude or Claude Code, and your Sunport’s agents work for you from there. ### Can I take my work elsewhere? Your agents run on open runtimes, their instructions are plain files, and you can copy files out and export your activity and charges at any time. A connected agent can leave and join another Sunport. ## Related guides - [Connect any agent](https://openorange.ai/connect-any-agent/) - [Agents and models](https://openorange.ai/runtime-adapters/) - [Model API](https://openorange.ai/model-api/) - [Sunport](https://openorange.ai/operator-layer/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=integrations&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=integrations&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/integrations.md) - [Deutsch](https://openorange.ai/de/integrations.md) - [Español](https://openorange.ai/es/integrations.md) - [Français](https://openorange.ai/fr/integrations.md) - [Italiano](https://openorange.ai/it/integrations.md) - [Português](https://openorange.ai/pt/integrations.md) - [Русский](https://openorange.ai/ru/integrations.md) - [Türkçe](https://openorange.ai/tr/integrations.md) - [Bahasa Indonesia](https://openorange.ai/id/integrations.md) - [Tiếng Việt](https://openorange.ai/vi/integrations.md) - [हिन्दी](https://openorange.ai/hi/integrations.md) - [ไทย](https://openorange.ai/th/integrations.md) - [日本語](https://openorange.ai/ja/integrations.md) - [한국어](https://openorange.ai/ko/integrations.md) - [简体中文](https://openorange.ai/zh/integrations.md) --- # One model API for your apps and tools Canonical URL: https://openorange.ai/model-api/ Updated: 2026-10-25 An OpenAI- and Anthropic-compatible model API in your OpenOrange Sunport. Use Claude Code, the SDKs, or your own app, with keys and usage in one place. Your apps and tools use the same models as your agents, on the same bill. ## Key capabilities - OpenAI- and Anthropic-compatible endpoints, so Claude Code and the SDKs work. - Every call shows its key, model, tokens, and cost. - No OpenOrange fee on calls with your own provider keys. ## Use the clients you already have Point any OpenAI- or Anthropic-compatible client at your Sunport, Claude Code and the SDKs included. Embeddings, transcription, speech and image models work too when your Sunport offers them. ## Usage and billing in one place API calls appear on the same Requests and Billing pages as your agents’ calls, and OpenOrange adds no fee to calls on your own provider keys. ## The same models as your agents Your apps, agents and coding sessions share one set of models, whether we provide them or you bring your own keys. ## Questions ### Does Claude Code work with OpenOrange? Yes. Create an API key in your Sunport and point Claude Code at your Sunport’s Anthropic-compatible endpoint. Its calls appear in your usage like any other. ### Which models can I call? Over 150 models from OpenAI, Anthropic, Google, xAI, DeepSeek, Moonshot and Z.ai, open families such as Qwen, Mistral and Llama, and any provider keys you connect. A key can be limited to the models you select. ## Related guides - [Agents and models](https://openorange.ai/runtime-adapters/) - [Costs and usage](https://openorange.ai/model-usage-accounting/) - [OpenOrange Code](https://openorange.ai/openorange-code/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=model-api&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=model-api&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/model-api.md) - [Deutsch](https://openorange.ai/de/model-api.md) - [Español](https://openorange.ai/es/model-api.md) - [Français](https://openorange.ai/fr/model-api.md) - [Italiano](https://openorange.ai/it/model-api.md) - [Português](https://openorange.ai/pt/model-api.md) - [Русский](https://openorange.ai/ru/model-api.md) - [Türkçe](https://openorange.ai/tr/model-api.md) - [Bahasa Indonesia](https://openorange.ai/id/model-api.md) - [Tiếng Việt](https://openorange.ai/vi/model-api.md) - [हिन्दी](https://openorange.ai/hi/model-api.md) - [ไทย](https://openorange.ai/th/model-api.md) - [日本語](https://openorange.ai/ja/model-api.md) - [한국어](https://openorange.ai/ko/model-api.md) - [简体中文](https://openorange.ai/zh/model-api.md) --- # Compute for agents and your own apps Canonical URL: https://openorange.ai/compute/ Updated: 2026-10-25 Create virtual machines for agents and your own apps, publish apps at their own HTTPS addresses, and see each machine’s running time next to your other usage. Machines come with your Sunport, and their running time shows up on the same bill as your models. ## Key capabilities - A dedicated virtual machine for every agent. - Machines for your own apps, each at its own HTTPS address. - Running time shown with the rest of your usage. ## A machine for every agent OpenClaw and Hermes agents each run on their own dedicated virtual machine, so one agent’s tools, files, and load stay apart from another’s. Deleting an agent deletes its machine too. ## Run your own apps Create machines for your own apps from the sizes and locations your Sunport offers, and publish an app at its own HTTPS address under your Sunport’s domain, with the certificate handled for you. ## Questions ### Who can create machines? Administrators. Members use the agents and projects that run on them, while infrastructure controls stay with the people managing the Sunport. ### Can I use my own domain? Sunports created by sign-up publish apps at an HTTPS address under the Sunport’s domain. For your own domain on an enterprise deployment, contact contact@openorange.ai. ## Related guides - [AI infrastructure](https://openorange.ai/ai-infrastructure/) - [Agents and models](https://openorange.ai/runtime-adapters/) - [Costs and usage](https://openorange.ai/model-usage-accounting/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=compute&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=compute&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/compute.md) - [Deutsch](https://openorange.ai/de/compute.md) - [Español](https://openorange.ai/es/compute.md) - [Français](https://openorange.ai/fr/compute.md) - [Italiano](https://openorange.ai/it/compute.md) - [Português](https://openorange.ai/pt/compute.md) - [Русский](https://openorange.ai/ru/compute.md) - [Türkçe](https://openorange.ai/tr/compute.md) - [Bahasa Indonesia](https://openorange.ai/id/compute.md) - [Tiếng Việt](https://openorange.ai/vi/compute.md) - [हिन्दी](https://openorange.ai/hi/compute.md) - [ไทย](https://openorange.ai/th/compute.md) - [日本語](https://openorange.ai/ja/compute.md) - [한국어](https://openorange.ai/ko/compute.md) - [简体中文](https://openorange.ai/zh/compute.md) --- # Security built into every Sunport Canonical URL: https://openorange.ai/security/ Updated: 2026-10-25 New OpenOrange Sunports run on sealed servers that install only signed releases, encrypt content at rest, and log every maintenance opening. Each customer’s Sunport runs on its own virtual server, sealed after install and updated only from signed releases. ## Key capabilities - A sealed server for every new Sunport. - Only signed releases install on new Sunports. - Maintenance access opened from your dashboard and logged. - Chat history, uploaded files, and model request content encrypted at rest. ## Sealed servers Once a new Sunport is installed, its server is sealed and updates itself only from signed releases. ## You open maintenance access When maintenance needs the server, an administrator opens access for a set time and can end it early, and every opening is logged. ## What we receive Settings › Data sharing shows what your Sunport sends us to run, bill and update it, and lets an administrator share more, such as redacted server logs. ## Questions ### How long does maintenance access stay open? When maintenance needs the server, an administrator opens access from the dashboard for one hour (the default), 24 hours or until closed, and can end it early. Each opening is logged. ### Are my logs encrypted? Yes, on new Sunports: chat history, uploaded files, and model request content are encrypted at rest. Model providers you use still see the content they process. ## Related guides - [Privacy and audit](https://openorange.ai/privacy-and-audit/) - [Sunport](https://openorange.ai/operator-layer/) - [Compute](https://openorange.ai/compute/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=security&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=security&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/security.md) - [Deutsch](https://openorange.ai/de/security.md) - [Español](https://openorange.ai/es/security.md) - [Français](https://openorange.ai/fr/security.md) - [Italiano](https://openorange.ai/it/security.md) - [Português](https://openorange.ai/pt/security.md) - [Русский](https://openorange.ai/ru/security.md) - [Türkçe](https://openorange.ai/tr/security.md) - [Bahasa Indonesia](https://openorange.ai/id/security.md) - [Tiếng Việt](https://openorange.ai/vi/security.md) - [हिन्दी](https://openorange.ai/hi/security.md) - [ไทย](https://openorange.ai/th/security.md) - [日本語](https://openorange.ai/ja/security.md) - [한국어](https://openorange.ai/ko/security.md) - [简体中文](https://openorange.ai/zh/security.md) --- # AI cost tracking: understand your spend Canonical URL: https://openorange.ai/model-usage-accounting/ Updated: 2026-10-25 See what your agents use and what it costs. OpenOrange connects model and service activity to the people, projects, and work behind it. Every charge shows who and what it was for, so you know where the money goes. ## Key capabilities - One prepaid balance in dollars for models, services, and compute (100 JU = $1). - See which people, agents and models drive your costs. - No OpenOrange fee on calls with your own provider keys. ## One prepaid balance One balance in dollars covers models, services and machines for all your Sunports, and the Billing page shows where it goes. ## Know who spent what Follow every charge to the person, agent or API key behind it, and set daily, weekly and monthly limits. ## Your own provider keys Bring your own provider keys and see the same usage, with no OpenOrange fee. ## Questions ### Can I see what drives my costs? Yes. Connect costs to people, agents, API keys, models, requests, and connected services. Price history and usage details help explain spending over time. ### Can cache savings be explained? Yes. Observability shows each model’s prompt-cache hit rate, the tokens read from and written to the cache, and what caching saved. ## Related guides - [Pricing](https://openorange.ai/pricing/) - [AI activity](https://openorange.ai/ai-observability/) - [Model API](https://openorange.ai/model-api/) - [Sunport](https://openorange.ai/operator-layer/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=model-usage-accounting&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=model-usage-accounting&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/model-usage-accounting.md) - [Deutsch](https://openorange.ai/de/model-usage-accounting.md) - [Español](https://openorange.ai/es/model-usage-accounting.md) - [Français](https://openorange.ai/fr/model-usage-accounting.md) - [Italiano](https://openorange.ai/it/model-usage-accounting.md) - [Português](https://openorange.ai/pt/model-usage-accounting.md) - [Русский](https://openorange.ai/ru/model-usage-accounting.md) - [Türkçe](https://openorange.ai/tr/model-usage-accounting.md) - [Bahasa Indonesia](https://openorange.ai/id/model-usage-accounting.md) - [Tiếng Việt](https://openorange.ai/vi/model-usage-accounting.md) - [हिन्दी](https://openorange.ai/hi/model-usage-accounting.md) - [ไทย](https://openorange.ai/th/model-usage-accounting.md) - [日本語](https://openorange.ai/ja/model-usage-accounting.md) - [한국어](https://openorange.ai/ko/model-usage-accounting.md) - [简体中文](https://openorange.ai/zh/model-usage-accounting.md) --- # Audit logs and access control for your AI Canonical URL: https://openorange.ai/privacy-and-audit/ Updated: 2026-10-25 Decide who can do what, keep spending in check and see a record of every privileged action, in your own private Sunport. Your content stays in your own Sunport, encrypted at rest, and you choose how long requests are kept. ## Key capabilities - Roles, allow and deny permissions, and spending limits for every person and agent. - Every privileged action recorded in an audit log. - Maintenance access opened from your dashboard and logged. ## Your data stays in your Sunport Chats, uploads and request content stay in your Sunport, encrypted at rest on new Sunports, and you choose how long requests are kept and who may open them. ## Clem works within your permissions Clem acts with the permissions of the person who asked, and deny permissions narrow what anyone can do, Clem included. ## A record of every privileged action Privileged work is recorded in an audit log, with who asked and what ran. ## Questions ### Where are my requests kept? Request content stays in your Sunport, encrypted at rest. While a request is processed, the software handling it and the configured model provider see its content. ### Can sensitive actions be blocked? Yes. Deny permissions block specific actions for any person or agent, including Clem, and agents need a person’s approval before they run commands on a virtual machine. ## Related guides - [Security](https://openorange.ai/security/) - [Sunport](https://openorange.ai/operator-layer/) - [AI activity](https://openorange.ai/ai-observability/) - [Costs and usage](https://openorange.ai/model-usage-accounting/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=privacy-and-audit&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=privacy-and-audit&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/privacy-and-audit.md) - [Deutsch](https://openorange.ai/de/privacy-and-audit.md) - [Español](https://openorange.ai/es/privacy-and-audit.md) - [Français](https://openorange.ai/fr/privacy-and-audit.md) - [Italiano](https://openorange.ai/it/privacy-and-audit.md) - [Português](https://openorange.ai/pt/privacy-and-audit.md) - [Русский](https://openorange.ai/ru/privacy-and-audit.md) - [Türkçe](https://openorange.ai/tr/privacy-and-audit.md) - [Bahasa Indonesia](https://openorange.ai/id/privacy-and-audit.md) - [Tiếng Việt](https://openorange.ai/vi/privacy-and-audit.md) - [हिन्दी](https://openorange.ai/hi/privacy-and-audit.md) - [ไทย](https://openorange.ai/th/privacy-and-audit.md) - [日本語](https://openorange.ai/ja/privacy-and-audit.md) - [한국어](https://openorange.ai/ko/privacy-and-audit.md) - [简体中文](https://openorange.ai/zh/privacy-and-audit.md) --- # Managed OpenClaw hosting for teams Canonical URL: https://openorange.ai/openclaw-runtime/ Updated: 2026-10-25 We run OpenClaw agents for your team in your own Sunport, a private AI workspace on its own virtual server. Each agent gets its own machine and browser, and your people chat with it in the dashboard, Telegram, Slack or Mattermost. ![Astrid, an OpenClaw agent, answers in a Sunport’s chat with a summary of the week’s support conversations and draft release notes.](https://openorange.ai/screenshots/agent-chat-light-2048.webp) ## Share agents with your team Invite people as administrators or users and give each agent an owner. Clem, the AI admin, sets up agents and invites people when you ask in chat. ## Let it use websites for you Each OpenClaw agent has its own files, tools and browser. Sign it in to a website once, and it keeps that login for later work. ## Choose models and spending limits Pick from models by OpenAI, Anthropic, Google and more. Decide which ones each person may use and what they may spend, and give each agent a budget of its own. ## Chat in Telegram, Slack and Mattermost Your team works with each agent in the dashboard or through its own Telegram, Slack and Mattermost bots, in direct messages and group chats. ## Bring your own OpenClaw server Already run OpenClaw on your own server? One line there makes it a managed agent in your Sunport, and you can hand it back to your own setup later. ## Free to start Your Sunport starts on a free trial, no card needed. After that, you pay for the models we provide from a prepaid balance, and your own keys cost nothing on OpenOrange. ## Create your first agent Create an account and ask Clem for an OpenClaw agent, or choose New agent on the Agents page. Chat with it as soon as its machine is ready. ## Questions ### How is this different from single-user OpenClaw hosting? It’s built for a team. Agents run in your private AI workspace on its own virtual server, your people work with them in their chat apps, Clem sets things up from chat, and you set each person’s access and spending. ### Can I run other agents too? Yes. Create Hermes agents or coding agents such as Codex, Gemini CLI and OpenCode from the same dialog, and connect agents that run elsewhere, such as Claude or a coding agent on your laptop. ### Can I use my own model keys? Yes. Bring keys from OpenAI, Anthropic, Google, OpenRouter and more, and OpenOrange charges nothing for those calls. ## Related guides - [Hermes hosting](https://openorange.ai/hermes-agent-hosting/) - [Connect any agent](https://openorange.ai/connect-any-agent/) - [Telegram, Slack and Mattermost](https://openorange.ai/telegram-slack-mattermost-agents/) - [Pricing](https://openorange.ai/pricing/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=openclaw-runtime&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=openclaw-runtime&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/openclaw-runtime.md) - [Deutsch](https://openorange.ai/de/openclaw-runtime.md) - [Español](https://openorange.ai/es/openclaw-runtime.md) - [Français](https://openorange.ai/fr/openclaw-runtime.md) - [Italiano](https://openorange.ai/it/openclaw-runtime.md) - [Português](https://openorange.ai/pt/openclaw-runtime.md) - [Русский](https://openorange.ai/ru/openclaw-runtime.md) - [Türkçe](https://openorange.ai/tr/openclaw-runtime.md) - [Bahasa Indonesia](https://openorange.ai/id/openclaw-runtime.md) - [Tiếng Việt](https://openorange.ai/vi/openclaw-runtime.md) - [हिन्दी](https://openorange.ai/hi/openclaw-runtime.md) - [ไทย](https://openorange.ai/th/openclaw-runtime.md) - [日本語](https://openorange.ai/ja/openclaw-runtime.md) - [한국어](https://openorange.ai/ko/openclaw-runtime.md) - [简体中文](https://openorange.ai/zh/openclaw-runtime.md) --- # OpenOrange Code Canonical URL: https://openorange.ai/openorange-code/ Updated: 2026-10-25 Build software with coding agents in your private Sunport, alone or with your team, each session in its own sandbox. Turn an idea into working software: a coding agent writes the code, and teammates follow along or fork a session to try their own approach. ## Key capabilities - Each session runs in its own sandbox with Git, Node.js and Python ready. - Teammates watch a session or fork their own copy. - Every session’s model use and cost traced to the person who ran it. ## Build with a coding agent Choose a model, give the agent context and queue the next task, or fork to try a different approach. ## A sandbox for every session Each session runs in its own sandbox with Git, GitHub CLI, Node.js, Python and a browser, apart from everything else. ## Work as yourself on GitHub Connect your own GitHub account, and your work goes out under your name. ## From the browser, the terminal, or your own tools Start a task with the oo command line or the Code API, then pick it up in the browser. ## Questions ### Can multiple people work in the same project? Yes. Share a session so teammates can view it or fork their own private copy. The creator stays the only writer, forks keep their lineage, and a durable queue holds the next prompts. ### Is each session kept apart? Yes. Each session runs in its own sandbox, so one project’s tools and files stay apart from another’s. ## Related guides - [Sunport](https://openorange.ai/operator-layer/) - [Agents and models](https://openorange.ai/runtime-adapters/) - [Costs and usage](https://openorange.ai/model-usage-accounting/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=openorange-code&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=openorange-code&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/openorange-code.md) - [Deutsch](https://openorange.ai/de/openorange-code.md) - [Español](https://openorange.ai/es/openorange-code.md) - [Français](https://openorange.ai/fr/openorange-code.md) - [Italiano](https://openorange.ai/it/openorange-code.md) - [Português](https://openorange.ai/pt/openorange-code.md) - [Русский](https://openorange.ai/ru/openorange-code.md) - [Türkçe](https://openorange.ai/tr/openorange-code.md) - [Bahasa Indonesia](https://openorange.ai/id/openorange-code.md) - [Tiếng Việt](https://openorange.ai/vi/openorange-code.md) - [हिन्दी](https://openorange.ai/hi/openorange-code.md) - [ไทย](https://openorange.ai/th/openorange-code.md) - [日本語](https://openorange.ai/ja/openorange-code.md) - [한국어](https://openorange.ai/ko/openorange-code.md) - [简体中文](https://openorange.ai/zh/openorange-code.md) --- # Managed Hermes Agent hosting Canonical URL: https://openorange.ai/hermes-agent-hosting/ Updated: 2026-10-25 We run Hermes Agent, the open-source agent from Nous Research, in your own Sunport, a private AI workspace on its own virtual server. Each Hermes agent gets its own machine and browser and works alongside your OpenClaw agents. ## Run Hermes next to OpenClaw Pick Hermes or OpenClaw when you create an agent. Both share your Sunport’s models, people and spending limits. ## Watch it work in its own browser Open its browser from the dashboard and watch what it does, live. ## Switch models whenever you like Give each Hermes agent the models that suit its work, from the ones we provide or on your own keys. ## Work with it in your group chats Add Hermes to a Telegram, Slack or Mattermost group, and it answers your people right there. ## Start on a free trial No card needed. After that, the models we provide draw from your prepaid balance, and your own keys cost nothing on OpenOrange. ## Create a Hermes agent Create an account, then ask Clem for a Hermes agent or choose New agent and pick Hermes. Chat with it as soon as its machine is ready. ## Questions ### Is this the Hermes Agent from Nous Research? Yes, the open-source one, and we keep it up to date for you. ### Can Hermes and OpenClaw agents work together? Yes. Either one can send the other a task and get the result back. ## Related guides - [OpenClaw hosting](https://openorange.ai/openclaw-runtime/) - [Connect any agent](https://openorange.ai/connect-any-agent/) - [Telegram, Slack and Mattermost](https://openorange.ai/telegram-slack-mattermost-agents/) - [Pricing](https://openorange.ai/pricing/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=hermes-agent-hosting&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=hermes-agent-hosting&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/hermes-agent-hosting.md) - [Deutsch](https://openorange.ai/de/hermes-agent-hosting.md) - [Español](https://openorange.ai/es/hermes-agent-hosting.md) - [Français](https://openorange.ai/fr/hermes-agent-hosting.md) - [Italiano](https://openorange.ai/it/hermes-agent-hosting.md) - [Português](https://openorange.ai/pt/hermes-agent-hosting.md) - [Русский](https://openorange.ai/ru/hermes-agent-hosting.md) - [Türkçe](https://openorange.ai/tr/hermes-agent-hosting.md) - [Bahasa Indonesia](https://openorange.ai/id/hermes-agent-hosting.md) - [Tiếng Việt](https://openorange.ai/vi/hermes-agent-hosting.md) - [हिन्दी](https://openorange.ai/hi/hermes-agent-hosting.md) - [ไทย](https://openorange.ai/th/hermes-agent-hosting.md) - [日本語](https://openorange.ai/ja/hermes-agent-hosting.md) - [한국어](https://openorange.ai/ko/hermes-agent-hosting.md) - [简体中文](https://openorange.ai/zh/hermes-agent-hosting.md) --- # Connect Claude, ChatGPT or any AI agent to your team Canonical URL: https://openorange.ai/connect-any-agent/ Updated: 2026-10-25 Claude, ChatGPT, a coding agent on your laptop or a bot you wrote can join your Sunport, your team’s private AI workspace. It shows up next to your OpenClaw and Hermes agents, and they message each other by name. ![The Agents page of a Sunport: Clem, OpenClaw and Hermes agents, and connected agents such as Claude Duet, Grokbot, Muse and Dots, each with its owner and status.](https://openorange.ai/screenshots/agents-light-2368.webp) ## Bring the agents you already use Coding agents such as Claude Code, Codex, Gemini CLI or OpenCode, assistants such as Claude or ChatGPT, and bots of your own can all join. ## Ask your agents from ChatGPT or Claude Add OpenOrange as a connector in ChatGPT, Claude or Claude Code, then ask your Sunport’s agents for help and get the results back in the same chat. ## Connect your own agent from the dashboard Choose New agent, then Connect, and paste what the dashboard gives you into your agent. It joins your Sunport, and administrators can pause or disconnect it at any time. ## Let it ask your agents and Clem A connected agent can ask your OpenClaw and Hermes agents or Clem for help and get their answers back. It can also ask for a server of its own, which an administrator approves. ## Connecting is free If the agent uses your Sunport’s models, its calls draw from your prepaid balance, and calls on your own keys cost nothing on OpenOrange. ## Questions ### Which agents can connect? Agents that can run a command on Linux, macOS or Windows (with WSL), and any app that supports MCP connectors, such as Claude and ChatGPT. ## Related guides - [Integrations](https://openorange.ai/integrations/) - [OpenClaw hosting](https://openorange.ai/openclaw-runtime/) - [Hermes hosting](https://openorange.ai/hermes-agent-hosting/) - [Telegram, Slack and Mattermost](https://openorange.ai/telegram-slack-mattermost-agents/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=connect-any-agent&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=connect-any-agent&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/connect-any-agent.md) - [Deutsch](https://openorange.ai/de/connect-any-agent.md) - [Español](https://openorange.ai/es/connect-any-agent.md) - [Français](https://openorange.ai/fr/connect-any-agent.md) - [Italiano](https://openorange.ai/it/connect-any-agent.md) - [Português](https://openorange.ai/pt/connect-any-agent.md) - [Русский](https://openorange.ai/ru/connect-any-agent.md) - [Türkçe](https://openorange.ai/tr/connect-any-agent.md) - [Bahasa Indonesia](https://openorange.ai/id/connect-any-agent.md) - [Tiếng Việt](https://openorange.ai/vi/connect-any-agent.md) - [हिन्दी](https://openorange.ai/hi/connect-any-agent.md) - [ไทย](https://openorange.ai/th/connect-any-agent.md) - [日本語](https://openorange.ai/ja/connect-any-agent.md) - [한국어](https://openorange.ai/ko/connect-any-agent.md) - [简体中文](https://openorange.ai/zh/connect-any-agent.md) --- # AI agents in Telegram, Slack and Mattermost Canonical URL: https://openorange.ai/telegram-slack-mattermost-agents/ Updated: 2026-10-25 Every OpenClaw and Hermes agent in your Sunport, your private AI workspace, can have its own bot in Telegram, Slack and Mattermost, so your team works with it in the apps they already use. ![Users › Access requests: two teammates who messaged Leif’s Telegram bot wait for an administrator to approve them as a user or an admin.](https://openorange.ai/screenshots/access-requests-light-2048.webp) ## Give each agent its own bot People pick which agent to talk to, and newcomers chat once you let them in. Ask Clem, and it helps you connect a bot. ## Ask Clem in a group chat Mention Clem in a group chat, channel or thread, and it acts for the person who asked, once they confirm. ## Chat apps cost nothing extra A chat in Telegram, Slack or Mattermost costs the same as one in the dashboard. ## Add your first bot Create an account and an agent, then add its bot on the agent’s Channels tab. ## Questions ### Which chat apps can agents use? Telegram, Slack and Mattermost, plus web chat in the dashboard. ### Do chats in these apps count toward spending limits? Yes, once a person connects their account. Their allowed models and spending limits apply, and the usage shows under their name. ## Related guides - [OpenClaw hosting](https://openorange.ai/openclaw-runtime/) - [Hermes hosting](https://openorange.ai/hermes-agent-hosting/) - [Connect any agent](https://openorange.ai/connect-any-agent/) - [Pricing](https://openorange.ai/pricing/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=telegram-slack-mattermost-agents&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=telegram-slack-mattermost-agents&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/telegram-slack-mattermost-agents.md) - [Deutsch](https://openorange.ai/de/telegram-slack-mattermost-agents.md) - [Español](https://openorange.ai/es/telegram-slack-mattermost-agents.md) - [Français](https://openorange.ai/fr/telegram-slack-mattermost-agents.md) - [Italiano](https://openorange.ai/it/telegram-slack-mattermost-agents.md) - [Português](https://openorange.ai/pt/telegram-slack-mattermost-agents.md) - [Русский](https://openorange.ai/ru/telegram-slack-mattermost-agents.md) - [Türkçe](https://openorange.ai/tr/telegram-slack-mattermost-agents.md) - [Bahasa Indonesia](https://openorange.ai/id/telegram-slack-mattermost-agents.md) - [Tiếng Việt](https://openorange.ai/vi/telegram-slack-mattermost-agents.md) - [हिन्दी](https://openorange.ai/hi/telegram-slack-mattermost-agents.md) - [ไทย](https://openorange.ai/th/telegram-slack-mattermost-agents.md) - [日本語](https://openorange.ai/ja/telegram-slack-mattermost-agents.md) - [한국어](https://openorange.ai/ko/telegram-slack-mattermost-agents.md) - [简体中文](https://openorange.ai/zh/telegram-slack-mattermost-agents.md) --- # OpenOrange pricing Canonical URL: https://openorange.ai/pricing/ Updated: 2026-10-25 Sign up, and your Sunport, your own private AI workspace, starts on a free trial with no card needed. After that, the models we provide draw from one prepaid balance, and calls on your own keys cost nothing on OpenOrange. ## Start with a free trial The trial includes credit for the models we provide, so your agents can start working before you add funds. ## One prepaid balance Your organization has one balance in dollars, shared by its Sunports. Add funds in the OpenOrange app whenever you need. ## What you pay for Calls to the models we provide, each charged at its model’s price. ## Use your own keys at no charge Bring keys from OpenAI, Anthropic, Google, Mistral, DeepSeek, OpenRouter and more. Your provider bills you, and OpenOrange charges nothing for those calls. ## When the balance runs out The models we provide and web search pause until you add funds. Your Sunport, agents and machines keep running, and your own keys keep working. ## Set limits for each person Cap what each person and agent may spend, see usage by day and model, and get an alert when the balance runs low. ## Custom terms for your company We can agree pricing and billing terms for your company. Email contact@openorange.ai. ## Questions ### Do I need a card to start? No. Sign up with just your email address. ### What is a Juice Unit? The unit OpenOrange measures usage in. Your balance shows in dollars, with Juice Units beside them. ### What happens when the trial ends? Add funds to keep using the models we provide. ## Related guides - [Costs and usage](https://openorange.ai/model-usage-accounting/) - [OpenClaw hosting](https://openorange.ai/openclaw-runtime/) - [Hermes hosting](https://openorange.ai/hermes-agent-hosting/) - [Connect any agent](https://openorange.ai/connect-any-agent/) ## Access and contact OpenOrange is for individuals, teams and companies, and large companies already use it. Create an account to start a free trial in a private Sunport of your own. Companies can email us for custom terms. Start free: https://app.openorange.ai/sign-up?ref=pricing&cta=markdown Sign in: https://app.openorange.ai/sign-in?ref=pricing&cta=markdown Contact: contact@openorange.ai ## Available languages - [English](https://openorange.ai/pricing.md) - [Deutsch](https://openorange.ai/de/pricing.md) - [Español](https://openorange.ai/es/pricing.md) - [Français](https://openorange.ai/fr/pricing.md) - [Italiano](https://openorange.ai/it/pricing.md) - [Português](https://openorange.ai/pt/pricing.md) - [Русский](https://openorange.ai/ru/pricing.md) - [Türkçe](https://openorange.ai/tr/pricing.md) - [Bahasa Indonesia](https://openorange.ai/id/pricing.md) - [Tiếng Việt](https://openorange.ai/vi/pricing.md) - [हिन्दी](https://openorange.ai/hi/pricing.md) - [ไทย](https://openorange.ai/th/pricing.md) - [日本語](https://openorange.ai/ja/pricing.md) - [한국어](https://openorange.ai/ko/pricing.md) - [简体中文](https://openorange.ai/zh/pricing.md)