langgenius/dify
langgenius
139.3k
Production-ready platform for agentic workflow development.
21.8k|TypeScript
NOASSERTIONaigptllm
Automation
Workflow engines and agent systems for AI processes that need more than chat.
Use this page when the work involves steps, tools, triggers, approvals, or internal apps that need to run repeatedly.
App and workflow builders
Dify and n8n are useful when teams need a visual product surface, workflow orchestration, or internal AI apps.
Programmable agents
CrewAI fits cases where agents need to coordinate roles, call tools, or run task loops in code.
Why it works
App builders
Dify, Flowise, and Langflow help teams build AI apps, agents, and internal workflows without wiring every step from scratch.
Agent orchestration
CrewAI, LangGraph, CAMEL, Eliza, and Semantic Kernel fit when work needs roles, state, tools, or programmable task flow.
Workflow automation
The n8n starter kit is a practical entry point for self-hosted automation connected to AI actions.
Curated repositories
Agent and workflow platforms
crewAIInc/crewAI
crewAIInc
50k
Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.
6.9k|Python
MITagentsaiai-agents
n8n-io/self-hosted-ai-starter-kit
n8n-io
14.7k
The Self-hosted AI Starter Kit is an open-source template that quickly sets up a local AI environment. Curated by n8n, it provides essential tools for creating secure, self-hosted AI workflows.
3.7k|Unknown
Apache-2.0ailow-codeself-hosted
FlowiseAI/Flowise
FlowiseAI
52.3k
Build AI Agents, Visually
24.2k|TypeScript
NOASSERTIONartificial-intelligencechatgptlarge-language-models
langflow-ai/langflow
langflow-ai
147.4k
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
8.9k|Python
MITreact-flowchatgptlarge-language-models
langchain-ai/langgraph
langchain-ai
30.5k
Build resilient language agents as graphs.
5.2k|Python
MITagentsaiai-agents
camel-ai/camel
camel-ai
16.8k
🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org
1.9k|Python
Apache-2.0ai-societiesartificial-intelligencedeep-learning
elizaOS/eliza
elizaOS
18.2k
Autonomous agents for everyone
5.5k|TypeScript
MITagentagenticai
microsoft/semantic-kernel
microsoft
27.8k
Integrate cutting-edge LLM technology quickly and easily into your apps
4.6k|C#
MITaiartificial-intelligencellm
Selection guide
How to choose a self-hosted agent or workflow platform
Decide whether you need a visual app builder, a low-code automation system, or a developer framework for programmable multi-agent workflows.
Visual AI app and workflow builders
Dify, Flowise, and Langflow fit when teams need a self-hosted product surface for LLM apps, RAG, prompt workflows, agents, and observability.
n8n starter kit for local automation
Best fit when you want a Docker Compose starting point that connects n8n, Ollama, Qdrant, and PostgreSQL for AI workflows.
Code-first agent frameworks
CrewAI, LangGraph, CAMEL, Eliza, and Semantic Kernel fit when developers want to define roles, state, flows, and tool calls in code rather than a no-code builder.
Project fit
App builders, frameworks, and starter kits serve different jobs
Dify, Flowise, and Langflow are visual builders; CrewAI, LangGraph, CAMEL, Eliza, and Semantic Kernel are code-first frameworks; and the n8n starter kit is a practical local automation stack rather than a standalone AI product.
Do not compare them as identical tools
The useful comparison is whether the team needs a visual product surface, programmable agents, or workflow automation across services.
Workflow fit
Agents are not always the answer
If the job is repeatable and operational, a workflow builder may be easier to maintain than an autonomous agent loop. If the job needs reasoning, tools, and role coordination, a code-first agent framework can fit better.
Non-coders
Dify, Flowise, Langflow, and n8n are easier to explain to teams that need visual workflows and shared internal apps.
Developers
CrewAI, LangGraph, CAMEL, Eliza, and Semantic Kernel are more relevant when the team expects to write and maintain agent behavior in code.
Suggested additions
Strong candidates not yet in the registry
n8n
n8n-io/n8n
A mature self-hostable workflow automation platform with AI agent and tool nodes. Stronger as a primary workflow product than the starter kit alone, with fair-code licensing to review.
View repositoryAgno
agno-agi/agno
A code-first framework and runtime for building agentic systems. Formerly Phidata, and a strong candidate once it is present in the upstream registry.
View repositoryActivepieces
activepieces/activepieces
A self-hostable automation platform with AI agents, MCP integrations, and workflow automation. Strong fit for teams comparing n8n-style AI workflows.
View repositoryTaskingAI
TaskingAI/TaskingAI
An AI app and agent backend with assistants, tools, and retrieval features. Relevant, but project activity should be checked before promoting it to the main list.
View repositoryOpenHands
All-Hands-AI/OpenHands
A self-hostable agent platform with strong adoption. Best treated as suggested here because its core fit is developer and coding agents rather than general workflow automation.
View repositoryJulep
julep-ai/julep
A stateful agent workflow platform with memory, tools, branching, and long-running tasks. Useful candidate, but should stay suggested due to project-transition caveats.
View repositoryTracecat
TracecatHQ/tracecat
A self-hostable security automation platform for teams and AI agents. Strong niche workflow fit, but too security-specific for the main generic list.
View repositoryToolJet
ToolJet/ToolJet
A self-hostable internal app builder with AI app and agent positioning. Useful adjacent candidate, but broader low-code app building is its main job.
View repositoryByteChef
bytechefhq/bytechef
An AI-native low-code platform for API orchestration, workflow automation, and agent integrations. Promising but less established than the main workflow options.
View repositoryGiselle
giselles-ai/giselle
An open-source AI app builder for agentic workflows. Relevant to visual agent building, but newer and lower-adoption than Dify, Flowise, and Langflow.
View repositoryNodeTool
nodetool-ai/nodetool
A visual builder for AI workflows and agents with local-model integrations. Interesting local-first workflow candidate, but still smaller than the main curated projects.
View repositoryRelated pages
Keep browsing
Self-hosted ChatGPT alternatives
Chat interfaces and assistant apps you can run with local models, private endpoints, or your own hosted providers.
Local model runtimes and inference servers
Inference servers and local runtimes for serving models on your machine, server, or private cloud.
Self-hosted RAG tools
Knowledge-base apps, retrieval frameworks, and document pipelines for private data and production AI systems.
Vector databases and retrieval storage
Databases and search layers for embeddings, metadata filtering, persistence, and semantic retrieval.
AI developer tools
Coding assistants and repo-aware tools that can run locally or inside private development environments.
FAQ
Questions answered
How are agents different from chat apps?
Agents execute steps, call tools, and coordinate workflows. Chat apps provide the conversation layer; agents handle the automation layer.
When should I use a workflow builder instead?
Use a workflow builder when the job is repeatable, multi-step, and easier to maintain as a flow than as a prompt loop.
Is Dify better than n8n for AI workflows?
Dify is a better fit for building LLM apps, prompt workflows, RAG experiences, and AI product surfaces. Flowise and Langflow are better when teams want visual agent graphs. n8n is stronger when the work is broader automation across services and AI is one part of the flow.
Should I choose a visual builder or an agent framework?
Use a visual builder such as Dify, Flowise, or Langflow when non-specialists need to inspect and change the workflow. Use a framework such as CrewAI, LangGraph, CAMEL, Eliza, or Semantic Kernel when the agent behavior belongs in code and needs version control, tests, or deeper customization.