Frontier models
Orkas OpenAI Anthropic Google Gemini DeepSeek Zhipu GLM Moonshot / Kimi MiniMax Doubao OpenRouter
Your AI teamOpen sourceLocal-firstSpecialist agents
Research, documents, slides, videos, websites… build your business right here
For indie builders, one-person businesses, and small teams: direct a team of AI specialists to deliver the work. Start free with managed official models, or connect your own provider.
Starting your download Didn’t start, or need another platform? Open the download page. Didn’t start or was blocked? Try another browser or adjust system security; for other platforms, open the download page.
Download the app and start free. Official models work out of the box, or connect your own model provider and local agents.
Thousandsindividuals and small teams using Orkas
94ready agents and skills
1.4k+GitHub stars
Weeklyreleases without interruption
Frontier agents
Claude Code Codex OpenClaw OpenCode Hermes
Real deliverables
Describe what you want to accomplish. The AI team builds the result.
The AI team handles research, writing, office, operations, and much more, then delivers files you can use immediately or keep editing.
Choose a market
Choose the next market with evidence, not instinct
Bring customer interviews, usage data, competitor material, and current sources into one evidence trail. Contradictions and gaps stay visible in a recommendation your team can review.
brief.docxevidence.csvreview.pdf
Explore the full workflow Product launch campaign
Turn one approved product brief into an article, images, and video
Keep the same product facts, CTA, and creative direction across the launch article, social image, and reviewable 60-second video.
article.mdsocial.pnglaunch.mp4
Explore the full workflow 
Monthly business review
Reconcile the numbers, then walk into the review ready
Reconcile the sales workbook and meeting notes first, then use the same checked figures to draft the report copy, update the analysis workbook, and build the review deck.
report.mdanalysis.xlsxreview.pptx
Explore the full workflow Every Monday at 08:30
Set it once. Monday’s team update is prepared automatically
On schedule, automatically read last week’s tasks, messages, and meeting notes, then produce the team update, refresh project notes, and list follow-up owners and due dates for your review.
weekly-update.mdteam-notes.mdfollow-ups.csv
Explore the full workflow Core strengths
Professional and reliable, secure and controlled, open and extensible
Three foundations make an AI team useful for real work, not just impressive in a demo.
01
Professional and reliable
Clear ownership from planning to handoff, with reusable working knowledge after every run.
Commander scheduling
Turn a goal into an executable plan, then route, track, and verify the work.
Understand the goalBuild the planAssign and verify
Specialist teamwork
Each specialist owns a stage, while the result stays on one coordinated task line.
Deep researchSearch and growthVideo production
Reusable capabilities
Keep useful agents, skills, and project outputs ready for the next task.
Custom agentsReusable skillsLocal assets
Memory and evolution
Remember preferences and facts, review each run, and improve future plans.
Preferences and factsRun reviewSelf-evolution
02
Secure and controlled
Keep workspace data local by default, protect credentials, define boundaries, and confirm critical actions.
Local storage
Chats, files, project history, and knowledge material stay on your device by default.
Chats and filesProject historyLocal library
Encrypted credentials
Protect API keys and sensitive settings locally instead of exposing plain text.
Encrypted keysLocal credentialsLess exposure
Access boundaries
Decide what an agent can read, call, or change for each kind of work.
Data scopeAvailable toolsEdit boundary
Critical action confirmation
Pause before risky operations so you can confirm the action and cost.
Risky operationsCost promptsHuman approval
03
Open and extensible
Use your models, external agents, and connected systems without locking your work into one platform.
Open-source code
Inspect the implementation, self-host what you need, and avoid a black box.
Auditable codeSelf-hostingMIT license
Flexible model access
Choose managed official models or connect your own supported provider.
Official modelsYour providerSwitch anytime
External agent access
Call supported local agents from the same workspace and conversation.
Coding agentsLocal agentsOne conversation
MCP and connectors
Connect data sources, tools, and community capabilities through supported integrations.
Data sourcesTool connectorsCommunity skills
Full comparison
Orkas vs others
Orkas is stronger when work needs a team, local context, reusable workflows, and control over model keys.
| Tool | Main use | Setup cost | Workspace scope | Coordination / reuse | Data / key control | Best fit |
|---|---|---|---|---|---|---|
| OrkasAI team desktop | Command an AI team across code, docs, research, data, operations, and more. | Desktop app, ready to use | Full workflow workspace | Commander + reusable sub-agents | Official models + BYO keys | You want complex work to become a reusable team workflow. |
| CodexCoding agent | Work on coding tasks inside a repository. | Client / CLI | Repository | Single coding agent session | Client/account based | You mainly need focused repository coding. |
| Claude CodeCoding agent | Work on deep coding tasks inside a repository. | Client / CLI | Repository | Single coding agent session | Client/account based | You mainly need focused repository coding. |
| ClineEditor coding agent | Autonomous coding inside VS Code. | VS Code plugin | Editor workspace | Single editor conversation | Configured in plugin | You want hands-on coding where you edit. |
| LangChainAgent framework | Build custom agent systems in code. | Engineering required | Whatever you build | You design orchestration | You implement control | You are building agent infrastructure. |
| OpenClawSelf-hosted assistant gateway | Run a personal assistant across channels, agents, and skills. | Self-hosted setup | Gateway + channels + workspaces | Agent routing + sub-agent runs | Self-operated | You want a channel-first, self-hosted personal assistant. |
| Hermes AgentAlways-on server agent | Run background and scheduled automations. | Server setup | Server workflows | Cron / scripts | Self-operated | You need unattended background automation. |
| DeepSeek HarnessComposable agent runtime | Build and run a plugin-composed agent for workspace tasks. | Node.js / npm developer preview | Workspace / custom runtime | Replaceable agent loop + subagents | Local config + provider keys | You want to compose or extend the agent runtime itself. |
FAQ
Frequently asked questions
What is Orkas?
Orkas is an open-source, local-first AI workforce. A super-powered Commander coordinates specialist agents to complete complex work together.
What can I build with Orkas?
Use Orkas as a coordinated AI team for complex work: research markets, turn findings into reports or slides, read PDFs with citations, fix bugs and open PRs, analyze data, create apps or videos, organize inbox and calendar work, connect external tools, and more, then turn repeated work into reusable agents and skills.
How is Orkas different from ChatGPT, Codex, Claude Code, or Cline?
ChatGPT is a hosted assistant, while Codex, Claude Code, and Cline focus on coding sessions. Orkas is an open-source, local-first AI workforce where a Commander coordinates specialist agents to complete complex work together.
Do I need to bring my own model key?
No. Orkas can provide managed official models that run through Orkas-managed model service. You can also connect your own providers through OAuth or API keys, including OpenAI, Claude, Gemini, DeepSeek, Kimi, and GLM; bring-your-own-provider calls go directly to that provider.
Will my files, history, or API keys be uploaded to Orkas?
By default, your configs, history, knowledge bases, and generated files stay on your machine. If you bring your own keys, they are stored locally and requests go directly to your provider. Orkas-managed models and optional cloud features use Orkas services only for those selected features.
Is Orkas free, and what would I pay for?
The open-source desktop edition is free to download and self-host. Bring-your-own-provider usage is billed by your own provider. Paid Orkas plans add hosted services such as managed models, sync, and cloud storage. Remote control and team collaboration are planned.
What are Orkas credits, and when do I need them?
Credits meter the Orkas-managed model service only. If you connect your own provider key, those calls go straight to that provider and are billed by it, so they use no credits. Paid plans include a monthly allowance — 5,000 on Lite, 10,000 on Pro, and 50,000 on Max — and you can also buy lifetime credit packs as a one-time purchase, which work exactly like monthly credits and are available whether or not you subscribe (monthly credits are spent first). Consumption follows each provider's official pricing: roughly 100–300 credits per million input tokens, 2–6 per million cached input tokens, and 200–600 per million output tokens, so repeated context costs far less than it looks.
Is Orkas open source, and where is the repository?
Yes. The Orkas desktop client is open source under the MIT license at github.com/Orkas-AI/Orkas. You can read the source, build it yourself, and audit exactly what the app sends over the network.
Can I self-host Orkas, and how does the open-source build differ from the official release?
Yes. The open-source desktop edition is free to download and self-host, and it runs without an Orkas account: you add your own provider keys and your conversations, files, and keys stay on your machine. The official release adds the optional hosted services — managed models, multi-device sync, and cloud storage — and those need an Orkas account. One thing worth being clear about: Orkas is a desktop client, not a server, so self-hosting means running the app on your own machine rather than deploying a server for a team.
Which platforms and interface languages does Orkas support?
The official download page provides macOS and Windows builds. The open-source repository also supports Linux, which you build yourself. The interface ships in English, Chinese, Japanese, and Portuguese, and you can switch language inside the app at any time.
Your AI team is ready
Install the desktop app, choose a managed official model or connect your own provider, and put an AI team to work in minutes.
Commander + specialist agents Local-first by default MIT open source