# Mantlecore — Complete Platform Reference > Mantlecore is the cloud for AI agents — it gives your AI its own computer. Instead of chatting with AI, you delegate to it: pick an AI agent, choose a model, and it's live in seconds in its own private workspace, where it can research, write, analyze data, automate tasks — and remember everything between sessions. LLM access is dual-mode: Managed (buy prepaid credits, pick any supported model, no API key needed) or BYOK (bring your own API keys from any provider). No code, no setup, no DevOps. Website: https://mantlecore.ai Blog: https://mantlecore.ai/blog --- ## What is Mantlecore? ChatGPT is a conversation — you ask, it answers. An AI agent on Mantlecore is a worker — you give it a task in plain English, and it actually does it: browses the web, writes and runs code, creates files, and remembers everything across sessions. Mantlecore's job is the computer that agent works on: you pick an agent, choose a plan, and your agent is live in seconds, with no servers, containers, networking, TLS certificates, or storage to manage. For developers (https://mantlecore.ai/dev): pick a framework, pick a model, click deploy — your agent runs in a VM-isolated environment with persistent storage and its own subdomain. No Docker on your laptop. Every agent runs in its own hardware-isolated virtual machine — a real security boundary, not a shared container. Key benefits: - Persistent storage — files, chats, memory, and settings survive restarts, updates, and version upgrades - Automatic HTTPS — every agent gets its own secure subdomain - Daily off-site backups to a separate data center, plus manual snapshots with one-click restore - Managed LLM credits or bring-your-own API keys — every new account starts with $2 of free AI credit - 7-day free trial on the Starter plan — no charge today --- ## Supported AI Agent Frameworks ### Agent Zero - **Category:** Autonomous Agent - **What it does:** Agent Zero is a general-purpose AI agent framework with full access to its own Linux environment — it writes and executes code, installs packages, browses the web, manages files, and interacts with external APIs. It can delegate subtasks to subordinate agents, create its own tools, and maintains persistent memory across sessions. Supports multiple LLM providers including OpenAI, Anthropic, Google, and local models. - **Key features:** Full Linux system access — write & execute code, install packages; Autonomous web browsing and research; Persistent memory across conversations; Multi-LLM support (OpenAI, Anthropic, Google, local models); File management and external API interaction; Multi-agent delegation — splits work across subordinate agents - **Best for:** Automated code generation and debugging; System administration and DevOps tasks; Data processing and analysis pipelines; Web scraping and research automation; File conversion and batch processing - **License:** MIT - **GitHub:** https://github.com/agent0ai/agent-zero - **Website:** https://agent-zero.ai - **Deploy on Mantlecore:** https://mantlecore.ai/agents/agent-zero ### Hermes Agent - **Category:** Personal Assistant - **What it does:** Hermes is an open-source personal AI agent by Nous Research, designed to be your agent across every platform you already use. Bridges 20 messaging platforms (Slack, Telegram, WhatsApp, Signal, Email, Discord, Teams, Google Chat, and more) through a single gateway. Persistent identity via SOUL.md, MEMORY.md, and USER.md — your agent remembers conversations across sessions and grows a procedural-memory library of self-authored skills. Built-in curator periodically refines its own skills library. Multi-provider LLM support routed through our LiteLLM proxy — pick any model, swap anytime. Durable multi-agent kanban and persistent goal loops keep long-running work moving across sessions. - **Key features:** 20 messaging platforms — Slack, Telegram, WhatsApp, Signal, Email, Discord, Teams, more; Persistent identity — SOUL.md, MEMORY.md, USER.md that survive across sessions; Self-authored skills — agent writes its own procedures after complex tasks; Built-in curator refines the skills library on a weekly cadence; Multi-agent Kanban with heartbeat / zombie detection; Embedded TUI in dashboard — the canonical Hermes experience, in your browser; OpenAI-compatible API on a separate port for integrating with your IDE, scripts, or curl; Provider-agnostic — Anthropic, OpenAI, Google, Nous Portal, OpenRouter, all routed through LiteLLM - **Best for:** Personal AI assistant across Slack, Telegram, WhatsApp at once; Build a long-running agent that learns your team's preferences; Replace a virtual assistant — your Hermes pairs with your email and calendar; Autonomous research agent with persistent memory of prior work; Multi-agent orchestration via the built-in Kanban for complex projects - **License:** MIT - **GitHub:** https://github.com/NousResearch/hermes-agent - **Website:** https://hermes-agent.nousresearch.com - **Deploy on Mantlecore:** https://mantlecore.ai/agents/hermes --- ## Pricing Every plan includes all AI agents, the web interface, persistent storage, private workspace security, automatic daily backups, and one-click updates that roll back automatically on a failed health check. Larger plans add more processing power, more storage, longer backup history (7 days on Starter, 30 on the larger plans) and more manual snapshots (1, 3 and 5). Plan changes are upgrade-only. No free tier — new accounts get a 7-day free trial on the Starter plan. Annual billing saves 20%. ### Plans | Plan | vCPU | RAM | Storage | Monthly | Annual | |------|------|-----|---------|---------|--------| | Starter | 2 vCPU | 4 GB | 20 GB SSD | $19/mo | $15/mo ($182/yr) | | Builder | 4 vCPU | 8 GB | 30 GB SSD | $39/mo | $31/mo ($374/yr) | | Pro | 8 vCPU | 16 GB | 50 GB SSD | $79/mo | $63/mo ($758/yr) | **Starter** — Perfect for personal projects, learning, and automating everyday tasks. - All AI agents included - 20 GB SSD storage included - Web-based interface - Expandable storage add-ons - Daily off-site backups (7-day history) - 1 manual snapshot - Private workspace security - $2 free AI credit to start **Builder** — For professionals running research, analysis, and complex automations. - Everything in Starter, plus: - 2x CPU and memory - 30 GB SSD storage included - Expandable storage add-ons - 30-day backup history - 3 manual snapshots - Priority email support - $2 free AI credit to start **Pro** — For professionals with heavy workloads that need sustained performance. - Everything in Builder, plus: - 4x CPU and memory - 50 GB SSD storage included - Expandable storage add-ons - 30-day backup history - 5 manual snapshots - Dedicated support - $2 free AI credit to start ### Storage Add-On - $2/month per 10 GB block (or $20/year) - Up to 500 GB total per instance (plan base + add-ons) - Storage expands live — no restart needed ### AI Model Access Two options, switchable per agent: - **Managed:** buy prepaid credits on the platform, pick any supported model (Anthropic, OpenAI, Google, and more) from a dropdown, and start working — no API keys needed. Model usage is billed at real provider prices with no per-token markup; a small platform fee is applied on credit purchases and shown before checkout. Auto-recharge is available. - **BYOK (bring your own key):** configure your own API keys from any provider directly in your agent. Keys stay inside your isolated workspace and never touch the platform. --- ## Security & Data - Every agent runs in its own hardware-isolated virtual machine (microVM), completely separated from other users - Automatic HTTPS with per-agent TLS certificates - Daily off-site backups to a separate data center, plus manual snapshots with one-click restore - Full terminal access to your VM, so you can copy files out at any time; deleting an instance removes its persistent volumes, and account deletion is by request --- ## FAQ **Q: Is there a free trial?** A: Yes. New accounts get a 7-day free trial on the Starter plan — full access, no charge today. We save your card so your agent keeps running after the trial; cancel anytime before it ends and you won’t be charged. **Q: What’s included in every plan?** A: Every plan includes all AI agents, the web interface, persistent storage, private workspace security, automated daily backups, and one-click updates that roll back automatically if a new version fails its health check. Larger plans add more processing power, more storage, longer backup history, and more manual snapshots. **Q: How does AI model access work?** A: Two options. Managed: buy prepaid credits on our platform, pick any supported model from a dropdown, and start working — no API keys needed. Or bring your own API keys from any provider. Every new account starts with $2 of free AI credit, so you can try managed mode without setting up keys. **Q: Are AI model credits included in my plan?** A: New accounts get $2 of free AI credit to start. Beyond that, your plan covers the workspace where your agent runs, and AI model usage is separate — top up with prepaid credits or use your own API keys. This keeps pricing simple and predictable. **Q: Can I change plans later?** A: You can upgrade anytime from your dashboard, and it applies to your running agent immediately — no rebuild and no downtime. Downgrades aren’t supported yet, so moving to a smaller plan means starting a new instance on it. **Q: Do I need technical skills to use this?** A: No. If you can type a message in a chat window, you can use Mantlecore. You describe what you want, and the agent does it. **Q: Is my data safe?** A: Yes. Every agent runs in its own hardware-isolated virtual machine — a real security boundary, not a shared container. Your files, conversations, and workspace are completely separated from every other user, and every plan includes automated daily backups with one-click restore. **Q: How is this different from ChatGPT?** A: ChatGPT is a conversation — you ask, it answers. An AI agent on Mantlecore is a worker — you give it a task, and it actually does it. It browses the web, writes and runs code, creates files, and remembers everything across sessions. **Q: Can I cancel anytime?** A: Yes. No contracts, no commitments. Cancel from your dashboard whenever you want. **Q: What happens to my data if I cancel?** A: Your workspace stays on its volume until you delete the instance, and you have full terminal access to copy anything out first. Deleting an instance removes its persistent volumes. To delete your account entirely, email us — see the Privacy Policy for the details. **Q: How does annual billing work?** A: Save 20% with annual billing. Switch between monthly and annual anytime. --- ## Links - Sign up: https://mantlecore.ai/signup - Pricing: https://mantlecore.ai/#pricing - All agents: https://mantlecore.ai/agents - For developers: https://mantlecore.ai/dev - Affiliate program: https://mantlecore.ai/affiliates - Blog: https://mantlecore.ai/blog - Contact: https://mantlecore.ai/contact - Privacy: https://mantlecore.ai/privacy - Terms: https://mantlecore.ai/terms