# The AI cloud you can run yourself — Hanzo AI > Build, deploy, and govern AI agents with unified access to models, MCP tools, memory, vector search, secure sandboxes, IAM, KMS, and audit logs. Open-source. Self-host or use the cloud. [V8 · Open Edition — the decentralized AI cloud](https://hanzo.ai/network) # The AI cloudyou can run yourself. Hanzo is a cloud you can run yourself. On hanzo.network, `hanzod` nodes start the same `cloud` binary we run in production — every capability behind one API, open source. Bring your own GPU or Kubernetes, or run all of it on your laptop. Connect any provider, watch usage and cost in one place, and let the router pick the model — saving up to 90% on compute. [Start building free](https://console.hanzo.ai)[Add compute, earn](https://hanzo.ai/network) [Explore capabilities](https://hanzo.ai/products)·[390+ open models](https://hanzo.ai/zen/models)·[One API](https://hanzo.ai/gateway)·[One binary](https://hanzo.ai/open-source)·[Bring your own GPU or K8s](https://hanzo.ai/cloud)·[Mine on any device](https://hanzo.ai/network)·[Free for public + OSS](https://hanzo.ai/pricing)·[Open source](https://hanzo.ai/open-source) [Backed by’17](https://www.techstars.com) console.hanzo.ai Agents Models Vector Gateway IAM Audit Agents running 12 Requests / min 48.2k Models served 390+ Inference throughput $ hanzo dev "Build a RAG API for ./docs. Add /chat. Write tests. Deploy it." One console for the whole cloud — models, agents, vector search, gateways, IAM, and audit. Run it in our cloud, on your own infrastructure, or on the network, with a complete audit trail. ## Backed by builders. Runs where you already run. Hanzo AI is a Techstars company. The platform deploys on the clouds and silicon you already trust. Backed by Runs on ## One API for all of it. Everything the cloud does is one route — `api.hanzo.ai/v1/`. Same key and same base URL for every one of them. [### AI Models, agents, and inference — open weights, one API.](https://hanzo.ai/products/ai)[### Compute GPUs, machines, and functions that scale to zero.](https://hanzo.ai/products/compute)[### Data Every persistence primitive — at rest and in motion.](https://hanzo.ai/products/data)[### Network Connect, route, and protect every service.](https://hanzo.ai/products/network)[### Security Identity, keys, and audit for the whole cloud.](https://hanzo.ai/products/security)[### Dev Clients, SDKs, and keys for the whole API.](https://hanzo.ai/products/dev)[### Infrastructure Source to production as declared state.](https://hanzo.ai/products/infrastructure)[### Observe Logs, metrics, traces, and cost in one pane.](https://hanzo.ai/products/observe)[### Web3 On-chain infrastructure for the cloud.](https://hanzo.ai/products/web3)[### Apps Production apps built on the cloud.](https://hanzo.ai/products/apps) Why Hanzo ## AI agents need infrastructure, not just prompts. A good demo can autocomplete a function. A production agent needs context, permissions, tools, memory, tests, secrets, sandboxes, and a way for humans to approve what matters. Hanzo gives agents the operating environment they need to do real work safely. ### Policy by default Permissions, approvals, and rate limits are enforced at the platform — not bolted on per agent. ### Evidence over vibes Every agent run produces a diff, a log, and a test result. Decisions are reviewable, not guessed. ### Persistent execution Long-running sandboxes install dependencies, refactor large codebases, and iterate until tests pass. ### Human control Plan-only, implement, test, deploy — pick the autonomy level. Humans approve what matters. ## Race agents. Ship the winner. Hanzo can run multiple models or agents against the same task in parallel, then compare results using tests, code quality, performance, and policy checks. GPT, Claude, Gemini, Mistral, Groq, xAI, DeepSeek, Qwen, Hugging Face, OpenRouter, Replicate, Together, Fireworks, and local models can all participate in the same workflow. ProvidersGPTClaudeGeminiMistralGroqxAIDeepSeekQwenHugging FaceOpenRouterReplicateTogetherFireworkslocal ### Parallel attempts Run multiple models or agents on the same task simultaneously in isolated sandboxes. ### Automatic scoring Score every attempt on tests, code quality, performance, and policy checks. ### Best patch wins Promote the winning diff; archive every attempt with full audit trail. ## Works where developers already work. Hanzo lives in your terminal, your IDE, your browser, and your API — sharing one brain across them all. ### Terminal Install with curl. Drive Hanzo Dev from the shell. Pipe agents into your build. ### VS Code & JetBrains First-class extension surface for editing, refactoring, and reviewing agent diffs. ### Browser Chrome, Firefox, and Safari extensions for in-context agent assistance. ### API OpenAI-compatible HTTP API for models, tools, memory, and agent execution. hanzo dev hanzo dev "Fix rate limiting. Add metrics. Add tests. Open a PR." ✓ Plan created ✓ Files updated ✓ Tests passing ✓ Pull request opened Hanzo Dev ## AI coding agent in your terminal. Plan, edit across files, run tests, and open pull requests — driven by your repo, not by guesswork. - Build APIs, scripts, components, and internal tools - Fix bugs with full codebase context - Add tests and CI workflows - Refactor large systems safely - Generate deployable services - Modernize legacy code [Read the Dev docs](https://docs.hanzo.ai/docs/dev) ## One shared brain for your AI tools. Hanzo connects your models, tools, memory, and history across every surface. Your terminal, IDE, browser, agents, and API calls all share the same context layer. ### Shared memory Context persists across every surface and every session. ### Vector search Semantic search over your entire codebase and knowledge base. ### Tool registry 260+ MCP tools available to every agent, IDE, and API call. ### Unified history Every prompt, every response, every tool call — searchable and replayable. ### Team sync Share context, memory, and projects across your team with org-scoped access. ## Secure remote execution for teams. Run long, parallel agent workloads on Hanzo Cloud or on your own Kubernetes — with isolation, quotas, and audit on by default. ### Parallel runs Execute work across repos and tasks simultaneously, with quotas per team. ### Shared workspaces Collaborate via shared sandboxed environments and shared memory. ### Quality gates Enforce policy, tests, and reviews before any change lands on main. ### Isolation Each run gets its own sandbox. Network, file, and tool egress are scoped. Hanzo API ## One API for models, tools, and agents. Hanzo provides a unified API for chat, model routing, tool calling, memory, and agent execution. It is compatible with existing OpenAI-style SDKs. Change the base URL and start routing through Hanzo. [View API docs](https://docs.hanzo.ai/docs) TypeScript Python Go cURL typescript import { Hanzo } from "@hanzo/ai"; const hanzo = new Hanzo({ apiKey: process.env.HANZO_API_KEY, }); const response = await hanzo.chat.completions.create({ model: "claude-sonnet-4-6", messages: [{ role: "user", content: "Hello!" }], }); console.log(response.choices[0].message.content); Native everywhere ## Hanzo AI, native in every tool you use. One API key, one identity. Sign in once with `@hanzo/iam` and the same models, tools, and memory follow you across your browser, editor, docs, and apps — all routed through the `@hanzo/ai` gateway. ### Browser [ChromeAsk, summarize, and automate any page from the toolbar.Install](https://github.com/hanzoai/extension/releases/latest) [EEdgeThe Hanzo sidebar and page actions, native in Edge.Install](https://github.com/hanzoai/extension/releases/latest) [FirefoxChat, extract, and act on tabs without leaving Firefox.Install](https://github.com/hanzoai/extension/releases/latest) [SafariHanzo assist and automation on macOS and iOS Safari.Install](https://github.com/hanzoai/extension/releases/latest) ### IDEs [VVS CodeAgent, chat, and inline edits inside VS Code.Install](https://github.com/hanzoai/extension/releases/latest) [CursorBring Hanzo models and tools into Cursor.Install](https://github.com/hanzoai/extension/releases/latest) [WindsurfRun Hanzo agents alongside the Windsurf editor.Install](https://github.com/hanzoai/extension/releases/latest) [AntigravityHanzo agents and context inside Antigravity.Install](https://github.com/hanzoai/extension/releases/latest) [JetBrainsIntelliJ, PyCharm, GoLand, and the full JetBrains suite.Install](https://github.com/hanzoai/extension/releases/latest) ### Docs & Office [WWord · Excel · PowerPointDraft, analyze, and build slides across Microsoft 365.Install](https://github.com/hanzoai/extension/releases/latest) [OOutlookTriage, draft, and summarize mail in Outlook.Install](https://github.com/hanzoai/extension/releases/latest) [Google WorkspaceDocs, Sheets, Slides, and Gmail with Hanzo built in.Install](https://github.com/hanzoai/extension/releases/latest) [PPDFRead, summarize, and extract from any PDF.Install](https://github.com/hanzoai/extension/releases/latest) ### Design [FigmaGenerate, edit, and annotate designs in Figma.Install](https://github.com/hanzoai/extension/releases/latest) [SketchHanzo assist for Sketch documents on macOS.Install](https://github.com/hanzoai/extension/releases/latest) [CCanvaDraft copy and lay out designs inside Canva.Install](https://github.com/hanzoai/extension/releases/latest) ### Comms [SSlackChat with Hanzo and run agents from any channel.Connect](https://docs.hanzo.ai) [MMicrosoft TeamsSummarize threads and act on messages in Teams.Connect](https://docs.hanzo.ai) [Zoom · Google MeetLive notes, summaries, and actions in your meetings.Connect](https://docs.hanzo.ai) ### Business & CRM [SSalesforceDraft, enrich, and summarize records in Salesforce.Connect](https://docs.hanzo.ai) [DDocuSignDraft and review agreements before you sign.Connect](https://docs.hanzo.ai) [NotionWrite, summarize, and query your Notion workspace.Connect](https://docs.hanzo.ai) [HubSpotDraft outreach and summarize deals in HubSpot.Connect](https://docs.hanzo.ai) [ShopifyGenerate product copy and analyze your store.Connect](https://docs.hanzo.ai) [ZendeskDraft replies and summarize tickets in Zendesk.Connect](https://docs.hanzo.ai) ### Developer [GitHubReview PRs, triage issues, and ship from GitHub.Connect](https://docs.hanzo.ai) [GitLabReviews, pipelines, and issues with Hanzo in GitLab.Connect](https://docs.hanzo.ai) [JupyterAI cells, explanations, and fixes in notebooks.Install](https://github.com/hanzoai/extension/releases/latest) [RaycastHanzo commands and agents from the Raycast launcher.Install](https://github.com/hanzoai/extension/releases/latest) ### Desktop & AI clients [DDesktop appThe full Hanzo app for macOS, Windows, and Linux.Download](https://hanzo.ai/download) [Claude DesktopBring Hanzo tools and models into Claude Desktop.Connect](https://docs.hanzo.ai) ### Industry [EEpicSummarize charts and draft notes in Epic (healthcare).Connect](https://docs.hanzo.ai) [PProcoreTrack RFIs, docs, and schedules in Procore (construction).Connect](https://docs.hanzo.ai) [QuickBooksCategorize, reconcile, and report in QuickBooks (finance).Connect](https://docs.hanzo.ai) [CClioDraft, review, and summarize matters in Clio (legal).Connect](https://docs.hanzo.ai) [Read the install guide](https://docs.hanzo.ai)[Download the apps](https://hanzo.app/download) Universal AI connector ## Connect every AI account. One usage plane. Hanzo is the connective layer for AI — link your accounts once and see limits, credits, and spend everywhere Hanzo runs. ### Connect once Link OpenAI, Anthropic, Hanzo, and 50+ AI providers — plus GPU, cloud, and integration accounts — using logins you already have. No passwords stored. ### See unified usage Rate limits, resets, credits, and spend for every account, in one view — native in the dev CLI, Hanzo Desktop, hanzo.app, hanzo.chat, and console.hanzo.ai. ### Share with your org Add your team to share verified logins and route usage through Hanzo for org-wide analytics and per-member reporting. 1% fee on routed usage. [Connect your accounts](https://console.hanzo.ai/ai-accounts)[Read the docs](https://docs.hanzo.ai/docs/usage) ## Every provider. One cloud. Connect Claude, Gemini, Grok, Groq, DeepSeek, Bedrock — 57+ AI providers in all. Hanzo is the login and usage manager in front of every one: see limits, credits, and spend in one plane, then let the native router serve the most optimal model for each request. [BBYO endpoint](https://console.hanzo.ai/ai-accounts/accounts?provider=byo)[Abacus AI](https://console.hanzo.ai/ai-accounts/accounts?provider=abacus)[Alibaba](https://console.hanzo.ai/ai-accounts/accounts?provider=alibaba)[Alibaba Token Plan](https://console.hanzo.ai/ai-accounts/accounts?provider=alibabatokenplan)[Amp](https://console.hanzo.ai/ai-accounts/accounts?provider=amp)[Antigravity](https://console.hanzo.ai/ai-accounts/accounts?provider=antigravity)[Augment](https://console.hanzo.ai/ai-accounts/accounts?provider=augment)[AAzure OpenAI](https://console.hanzo.ai/ai-accounts/accounts?provider=azureopenai)[AWS Bedrock](https://console.hanzo.ai/ai-accounts/accounts?provider=bedrock)[Chutes](https://console.hanzo.ai/ai-accounts/accounts?provider=chutes)[Claude](https://console.hanzo.ai/ai-accounts/accounts?provider=claude)[ClawRouter](https://console.hanzo.ai/ai-accounts/accounts?provider=clawrouter)[Codebuff](https://console.hanzo.ai/ai-accounts/accounts?provider=codebuff)[Codex](https://console.hanzo.ai/ai-accounts/accounts?provider=codex)[Command Code](https://console.hanzo.ai/ai-accounts/accounts?provider=commandcode)[Copilot](https://console.hanzo.ai/ai-accounts/accounts?provider=copilot)[Crof](https://console.hanzo.ai/ai-accounts/accounts?provider=crof)[CrossModel](https://console.hanzo.ai/ai-accounts/accounts?provider=crossmodel)[Cursor](https://console.hanzo.ai/ai-accounts/accounts?provider=cursor)[Deepgram](https://console.hanzo.ai/ai-accounts/accounts?provider=deepgram)[DeepSeek](https://console.hanzo.ai/ai-accounts/accounts?provider=deepseek)[Devin](https://console.hanzo.ai/ai-accounts/accounts?provider=devin)[Doubao](https://console.hanzo.ai/ai-accounts/accounts?provider=doubao)[ElevenLabs](https://console.hanzo.ai/ai-accounts/accounts?provider=elevenlabs)[Droid](https://console.hanzo.ai/ai-accounts/accounts?provider=factory)[Gemini](https://console.hanzo.ai/ai-accounts/accounts?provider=gemini)[Grok](https://console.hanzo.ai/ai-accounts/accounts?provider=grok)[Groq](https://console.hanzo.ai/ai-accounts/accounts?provider=groq)[JetBrains AI](https://console.hanzo.ai/ai-accounts/accounts?provider=jetbrains)[Kilo](https://console.hanzo.ai/ai-accounts/accounts?provider=kilo)[Kimi](https://console.hanzo.ai/ai-accounts/accounts?provider=kimi)[Kimi K2 (unofficial)](https://console.hanzo.ai/ai-accounts/accounts?provider=kimik2)[Kiro](https://console.hanzo.ai/ai-accounts/accounts?provider=kiro)[LiteLLM](https://console.hanzo.ai/ai-accounts/accounts?provider=litellm)[LLM Proxy](https://console.hanzo.ai/ai-accounts/accounts?provider=llmproxy)[Manus](https://console.hanzo.ai/ai-accounts/accounts?provider=manus)[Xiaomi MiMo](https://console.hanzo.ai/ai-accounts/accounts?provider=mimo)[MiniMax](https://console.hanzo.ai/ai-accounts/accounts?provider=minimax)[Mistral](https://console.hanzo.ai/ai-accounts/accounts?provider=mistral)[Moonshot / Kimi API](https://console.hanzo.ai/ai-accounts/accounts?provider=moonshot)[Ollama](https://console.hanzo.ai/ai-accounts/accounts?provider=ollama)[OpenAI](https://console.hanzo.ai/ai-accounts/accounts?provider=openai)[OpenCode](https://console.hanzo.ai/ai-accounts/accounts?provider=opencode)[OpenCode Go](https://console.hanzo.ai/ai-accounts/accounts?provider=opencodego)[OpenRouter](https://console.hanzo.ai/ai-accounts/accounts?provider=openrouter)[Perplexity](https://console.hanzo.ai/ai-accounts/accounts?provider=perplexity)[Poe](https://console.hanzo.ai/ai-accounts/accounts?provider=poe)[Qoder](https://console.hanzo.ai/ai-accounts/accounts?provider=qoder)[Sakana AI](https://console.hanzo.ai/ai-accounts/accounts?provider=sakana)[StepFun](https://console.hanzo.ai/ai-accounts/accounts?provider=stepfun)[Synthetic](https://console.hanzo.ai/ai-accounts/accounts?provider=synthetic)[T3 Chat](https://console.hanzo.ai/ai-accounts/accounts?provider=t3chat)[Venice](https://console.hanzo.ai/ai-accounts/accounts?provider=venice)[Warp](https://console.hanzo.ai/ai-accounts/accounts?provider=warp)[Windsurf](https://console.hanzo.ai/ai-accounts/accounts?provider=windsurf)[z.ai](https://console.hanzo.ai/ai-accounts/accounts?provider=zai)[Zed](https://console.hanzo.ai/ai-accounts/accounts?provider=zed) Smart routing ## One model id. The right model every time. Hanzo Inference is the native router: it uses consensus to serve the most optimal model for each prompt — local zen, a cheap tier, or a frontier API — matched to the task and your SLO. Frontier quality without frontier bills: up to 90% lower AI spend, and up to 10× more efficient than a single-provider API. ### Send model: "auto" Point any OpenAI-compatible client at api.hanzo.ai /v1/chat/completions and set model:"auto". One model id — no per-call model picking, no vendor lock-in. ### Routed by task + SLO A built-in heuristic — upgradeable to the trained zen-router — sends each prompt to a local zen model, a cheap tier, or a frontier model, by difficulty and your latency/quality SLO. ### Billed as what served you You pay for the model that actually answered — not a flat premium. Every response carries an X-Routed-Model header, so routing is fully transparent and auditable. A typical workload, routed Most prompts are easy. Only the hard ones need a frontier model. ~89% lower API spend on this example mix 70%On-device zen simple prompts · ~$0 20%Cheap tier mid-difficulty 10%Frontier hardest prompts Illustrative, not a guarantee — savings depend on your prompt mix. On the public RouterBench data, oracle routing reaches 0.909 quality at roughly 13× lower cost than the best single model (replication smoke run). [Read the research](https://papers.hanzo.ai/hanzo-router). [Turn on smart routing](https://console.hanzo.ai/ai-accounts)[Read the docs](https://docs.hanzo.ai/docs/usage/routing) Open Source ## Open source, one binary. The whole cloud is a single static binary you can read, fork, and run. `curl -fsSL hanzo.sh | bash` and self-host it on your own GPU or Kubernetes — the same cloud we run in production, no lock-in. ### Core projects [hanzoai/hanzoThe complete AI platform for agents, tools, models, and deployment.](https://github.com/hanzoai/hanzo)[hanzoai/mcp260+ Model Context Protocol tools and servers.](https://github.com/hanzoai/mcp)[hanzoai/llmOne API for every model — OpenAI- and Anthropic-compatible.](https://github.com/hanzoai/llm)[hanzoai/devAI coding agent for terminal and IDE workflows.](https://github.com/hanzoai/dev)[hanzoai/agentMulti-agent orchestration framework.](https://github.com/hanzoai/agent)[hanzoai/evalsEvaluation harness for models, agents, and workflows.](https://github.com/hanzoai/evals)[hanzoai/uiReact components for AI applications.](https://github.com/hanzoai/ui) We don't just ship the platform — we research and maintain the Zen family, top open-weight models anyone can run, on-device or in the cloud. [View on GitHub](https://github.com/hanzoai)[Explore Zen models](https://hanzo.ai/zen)[Read the research](https://hanzo.blog) Hanzo Network ## The network is the cloud. hanzo.network is a decentralized network of `hanzod` nodes that spawn and power the same unified cloud binary we run in production. Bring any device, mine at market price, and run the whole cloud yourself — no single operator owns it. ### Same binary, everywhere hanzod nodes spawn the exact unified cloud binary we run in production. The network is the substrate; the cloud is one binary on top. ### Mine on any device Bring your own GPU or Kubernetes and earn at market price. Confidential workloads run inside NVIDIA TEE / Confidential Compute. ### Free for public + OSS Public and open-source workloads run free on public nodes. GitHub-for-compute — no gatekeeping, no lock-in. [Add compute, earn](https://hanzo.ai/node)[Explore the network](https://hanzo.ai/network)[View source](https://github.com/hanzoai) Pricing ## Start free. Scale on your terms. No credit card to get started. Move to Enterprise when you need higher limits or private deployment. Free ### $0/month Everything you need to build and ship your first AI agents. - Free to start — pay only for what you use - 60 requests/min, 100K tokens/min - OpenAI-compatible API - Chat, agents, and MCP tools - Community support - Pay-as-you-go after credit [Start building free](https://console.hanzo.ai)No credit card required Enterprise ### Custom For teams that need higher limits, private deployment, SLAs, dedicated support, compliance, and custom infrastructure. - Higher API and storage limits - Private cloud or on-prem deployment - SLAs and dedicated support - Audit report and BAA on request - Custom contracts and procurement - Solution architect engagement [Contact sales](https://hanzo.ai/contact/sales) [View detailed pricing](https://hanzo.ai/pricing) ## The AI cloud you can run yourself. Build on it in our cloud, add compute to the network and earn, or self-host the whole thing. One binary, three paths — pick yours. [Start building free](https://console.hanzo.ai)[Add compute, earn](https://hanzo.ai/node)[Self-host](https://docs.hanzo.ai/docs/network) terminal $ curl -fsSL hanzo.sh | bash $ hanzo dev "Build my first AI app"