AI Agents
14 tools
AI agent platforms let you build LLM-powered agents that execute multi-step tasks: support, content generation, data analysis, coding. Three types: low-code platforms (Dify, Coze), dev frameworks (LangChain, CrewAI, AutoGen) and vertical apps. Selection criteria: model support (GPT/Claude/domestic), self-hosting options, plugin ecosystem, and hidden token-based costs.
MetaGPT
AI Agents
Multi-agent framework simulating a software company
Public purchase / signup
FastGPT
AI Agents
Open-source knowledge-base QA platform
Public purchase / signup
Lindy
AI Agents
AI assistant that automates tasks and workflows.
Public purchase / signup
Devin
AI Agents
AI software engineer that autonomously writes and fixes code.
Manus
AI Agents
AI agent that turns ideas into actions.
Public purchase / signup
Relevance AI
AI Agents
Build and manage AI agents and workflows without code.
Request a quote
AutoGPT
AI Agents
Autonomous AI agent platform for building and deploying agents.
Public purchase / signup
Flowise
AI Agents
Build LLM apps and AI agents with a drag-and-drop interface.
Public purchase / signup
Caddi
AI Agents
Agent that builds agents by only showing your work only once Discussion | Link
Public purchase / signup
CrewAI
AI Agents
Multi-agent orchestration framework
Public purchase / signup
Dify
AI Agents
Open-source LLM app development platform
Public purchase / signup
Microsoft AutoGen
AI Agents
Multi-agent framework from Microsoft Research
Public purchase / signup
LangChain
AI Agents
The default framework for LLM apps
Public purchase / signup
Coze
AI Agents
ByteDance's no-code bot builder
Public purchase / signup
FAQ
Low-code platform or dev framework?
Business teams validating use cases: Dify/Coze (live in a day). Deep customization or embedding into your systems: LangChain/CrewAI. Many teams use both — prototype in low-code, productionize on a framework.
Can AI agents replace customer support?
They handle 60-80% of standard queries (returns, lookups), but complex complaints and emotional cases still need humans. Use a hybrid model: AI answers first, escalates when unsure.
What hidden costs do agent platforms have?
Token fees (thousands of dollars monthly at scale), vector DB storage, and ongoing prompt-engineering effort.