AI News HubLIVE
站內改寫4 分鐘閱讀

待翻譯:Show HN: Maritime, a platform for running AI agents for $1 a month

AI 服務暫時不可用,以下為來源摘要,待恢復後補全翻譯:Spin up OpenClaw, Hermes, or ZeroClaw in one command. spin up Run a thousand agents as easily as one.Run 100 agents as easily as one. Run thousands of customer-facing AI agents on Maritime with isolated microVMs, effort…

來源Hacker News AI作者: mariagorskikh1

AI 服務暫時不可用,以下為來源正文,待恢復後補全翻譯。

Spin up OpenClaw, Hermes, or ZeroClaw in one command. spin up Run a thousand agents as easily as one.Run 100 agents as easily as one. Run thousands of customer-facing AI agents on Maritime with isolated microVMs, effortless scaling, and fully managed infrastructure. Start freeBook a demo No card to deploy your first agent maritime.sh/agents/atlas Get this dashboard Developers building on Maritime work at these companies Govern d'Andorra One agent per customer. Provisioned from your backend. Every signup gets its own agent. One API call brings up a dedicated micro-VM: tagged with your IDs, holding its own secrets, sending you lifecycle webhooks. No shared workers. No customer data in another customer's VM. Read the build docs SDKs TypeScript on npm, Python on PyPI, REST underneath. Your IDs, our infra Tag agents with external IDs so your database stays the source of truth. Webhooks Know the moment any agent deploys, sleeps, wakes, or fails. CLI Fully non-interactive, so your own agents can provision agents. Not a shared runtime. A micro-VM per agent. One command to a running agent. Maritime detects your framework, builds an isolated micro-VM, and puts your agent live. No infra to wire up. Cheap: $1 per agent a month Run one agent or hundreds. Each VM is sized to its work and sleeps when idle, so you never pay for a server sitting warm. Convenient: one CLI, one SDK A real CLI, a typed SDK, a dashboard, and a library of templates to start from. No glue code. Any framework, any model Run OpenClaw, Hermes, ZeroClaw, or your own stack. Bring whatever model and tools you already use. Fast: live in seconds One command from your terminal to a running agent. It sleeps when idle and wakes the moment traffic arrives. Isolation your customers can be told about. Hard boundaries A micro-VM per agent with its own kernel, not threads in someone else's process. OPENAI_API_KEY**************** CRM_API_KEY************** DATABASE_URL************** ANTHROPIC_API_KEY************** GOOGLE_AUTH_SECRET**************** Encrypted secrets Credentials encrypted at rest, mounted only into the one VM they belong to. No lock-in Standard frameworks, SSH into any VM, exportable state, guided migrations in from E2B, LangGraph, and eight other platforms. The only platform designed to scale to thousands of agents. Maritime E2B Daytona Modal AWS EC2 Google Cloud DigitalOcean Railway Fly.io Monthly cost $1/mo flat $48-198/mo $48/mo $23-68/mo $8-15/mo $6-49/mo $6/mo $5-10/mo $6+/mo Setup time Minutes Minutes Minutes Minutes Hours Hours Minutes Minutes Medium No cold starts None None None Yes None Yes None None None Stateful containers Yes Yes Yes No Yes No Yes No Yes Sleep / wake Built-in Pause API Auto-stop No No No No No No Agent-first design Yes Yes Yes No No No No No No From zero to a running agent in about a minute. MARITIME Create one harness pick a framework, or bring your own code OpenClaw Hermes ZeroClaw CrewAI LangGraph Build your agent locally. Any framework, any code. Connect to Maritime any route in works GitHubgit push maritime main $CLImaritime deploy SDKnpm i maritime Dockerdocker push Over GitHub, the CLI, the SDK, or a Docker image. Maritime handles the rest agents1 awake asleep, freeisolated micro-VMs Isolation, sleep and wake, and scaling, per agent. ~1s Wake from sleep Memory, files, and secrets exactly where it left them. $0 Cold-start tax Streams and voice pass straight through. auto Capacity as you grow A launch-day spike is handled before you notice it. One platform, from your first agent to the millions you sell. Developers Ship one agent for a dollar. Build it with whatever framework and tools you already use. ›SSH into the machine, read every log ›Any LLM, any framework ›One agent or a whole swarm Deploy your first agent Startups Sell agents to your customers. Build one, add a few lines of code, and Maritime runs a copy for every customer who signs up. ›One isolated micro-VM per customer ›Optional LLM proxy in front ›A thousand overnight sign-ups, handled Book a demoSee the startup setup Enterprises Run Maritime on-prem. Install it on hardware you already own and host thousands of isolated agents. ›Your data stays on your infrastructure ›Massive density on existing servers ›Auto-scaling and isolation inside your network Talk to us Flat pricing. Nothing metered. How many agents will you run?7 Free$0 Agents included3 SupportDiscord community Start free Starter$20/mo Agents included20 Extra agentper agent / mo$1.50 Extra RAMper GB / agent / mo, 2 GB included, up to 8$3 Extra SSDper 5 GB / agent / mo, 5 GB included, up to 100$2 SupportDiscord community Get started Growth$100/mo Agents included100 Extra agentper agent / mo$1.25 Extra RAMper GB / agent / mo, 2 GB included, up to 8$2.50 Extra SSDper 5 GB / agent / mo, 5 GB included, up to 100$1.50 SupportPrivate Slack + your rep Get started Scale$500/mo Agents included500 Extra agentper agent / mo$1.00 Extra RAMper GB / agent / mo, 2 GB included, up to 8$2 Extra SSDper 5 GB / agent / mo, 5 GB included, up to 100$1 SupportPrivate Slack + your rep Get started EnterpriseCustom Agents includedCustom Extra agentper agent / moVolume Extra RAMper GB / agent / mo, 2 GB included, up to 8Custom Extra SSDper 5 GB / agent / mo, 5 GB included, up to 100Custom SupportDedicated Talk to us Free $0 Starter $20/mo Growth $100/mo Scale $500/mo Enterprise Custom Agents included 320100500Custom Extra agent per agent / mo –$1.50$1.25$1.00Volume Extra RAM per GB / agent / mo, 2 GB included, up to 8 –$3$2.50$2Custom Extra SSD per 5 GB / agent / mo, 5 GB included, up to 100 –$2$1.50$1Custom Support Discord communityDiscord communityPrivate Slack + your repPrivate Slack + your repDedicated Start freeGet startedGet startedGet startedTalk to us Every agent comes with 1 vCPU, 2 GB RAM, and 5 GB SSD, and wakes from sleep in about a second. Full pricing and FAQ Building an agent company? The Maritime startup program covers your infrastructure with hosting and LLM credit grants. See the startup program Deploy your first agent for $1. Start from a template and ship in about a minute. Scale when you're ready and only pay for what you run. Spin up 3 agents for freeBook a demo