lobehub logo

lobehub

Free

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effo

Model APIsFreeFree tier
Inputs: text, image, fileOutputs: text, image, video
Type
Open Source

About lobehub

LobeHub is an open-source AI platform that serves as your Chief Agent Operator, enabling you to hire, schedule, and manage a team of AI agents that work 24/7. Built on the Agent Harness—an OS-like runtime that orchestrates models, tools, and memory—LobeHub lets you use any large language model, connect plugins, and own your data. Originally launched as LobeChat, it has evolved into a comprehensive platform with over 700,000 GitHub stars, supporting personal assistants, enterprise workflows, and multi-agent collaboration. Features include long-horizon task management, agent memory, multimodal workflows, a marketplace for skills and MCP servers, and seamless integration with chat platforms like Discord. LobeHub is designed for individuals and teams who want to offload routine work to AI while maintaining full control and transparency.

Key Features

Agent Harness runtime: OS-like orchestration between models, tools, and memory
Long-horizon task management: hires, schedules, and reports on AI agents across projects
Multi-agent collaboration: parallel collaboration, auto team formation, iterative improvement
Agent builder with one-sentence setup: auto-configures roles, skills, and behaviors
Marketplace with 332,955+ skills and 79,961+ MCP servers
Multimodal workflow across Pages, Schedules, Projects, and a shared Workspace
Personal memory: builds a clear understanding of user, continual learning, adaptive behavior
White-box memory: structured, editable memory for transparency
IM Gateway: lets agents operate where users already chat (e.g., Discord)
Supports any LLM (DeepSeek, Claude, GPT, Grok, Gemini, Qwen, etc.)

Pros & Cons

Pros
  • Fully open-source with a large community (700,000+ GitHub stars)
  • Multi-model support: run any LLM including local models with one click
  • Agent Harness provides rich orchestration (tools, memory, sub-agents) out of the box
  • White-box memory and editable agent behaviors give transparency and control
  • Free tier available with 500,000 credits per month; no credit card required to start
  • Cloud and self-hosted options for teams and enterprises
Cons
  • Advanced models and higher usage require paid credit packs; free tier limited
  • Complexity for non-technical users: building and managing agents may have a learning curve
  • Cloud version dependency on credits; self-hosting requires technical setup
  • Pricing page shows multiple inconsistent plans, potentially confusing
  • Some features (e.g., video generation) only available in paid plans

Best For

Automating issue management: sweep 500 open issues with dispatched agentsCustomer support: whisper on Discord with automated agent responsesContent creation: write and refine pages with multiple agents sharing contextPersonal assistant: schedule tasks, manage projects, and offload routine workEnterprise workflow automation: deploy private AI teams with custom models and integrationsResearch and analysis: use multimodal agents for data gathering and report generation

Alternatives to lobehub

FAQ

What is LobeHub?
LobeHub is an open-source AI platform that lets you hire, schedule, and manage a team of AI agents 24/7. It provides an Agent Harness runtime to orchestrate models, tools, and memory, supporting any LLM and multi-agent collaboration.
Is LobeHub free?
Yes, LobeHub offers a free plan with 500,000 credits per month for basic model usage (e.g., DeepSeek V4 Pro). Paid plans (Starter, Premium, Ultimate) start at $9.9–$12.9 per month and unlock more credits, advanced models, and features like vector storage and priority support.
What models does LobeHub support?
LobeHub supports a wide range of models including DeepSeek V4 Pro/Flash, Claude Sonnet 5, GPT-5.6 Sol/Terra, Grok 4.5, Gemini 3.5 Flash, Qwen3.7, and many others. You can also connect your own private models.
Can I self-host LobeHub?
Yes, LobeHub offers an Enterprise Edition for teams that need private deployment, custom solutions, commercial license, brand theming, and user management. Self-hosting is supported.
How do agents work in LobeHub?
Agents are built using the Agent Harness, which handles prompt presets, tool calling, lifecycle hooks, planning, filesystem access, and sub-agent management. You can create agents with one sentence—roles, skills, and behaviors are auto-configured. Agents can collaborate in groups, work across Pages and Projects, and learn from user interactions.