Category
SDKs and orchestration layers for building AI applications, workflows, and autonomous agents.
20 tools in this category
LangChain
A framework for building LLM applications, agent workflows, retrieval systems, and tool-calling chains.
LangChain is used to orchestrate prompts, tools, memory, retrieval, and multi-step workflows when developers want to build AI applications beyond a single model call.
Vercel
A TypeScript SDK for building streaming AI interfaces and model-powered web applications.
Vercel AI SDK is used to add chat, streaming responses, structured generation, and multi-provider model support to web apps built with modern JavaScript frameworks.
Agno
A framework for building agents that coordinate tools, memory, and workflows across different model providers.
Agno is used to assemble agentic systems, tool-calling flows, and internal assistants that need more structured orchestration than raw prompts.
Microsoft
A Microsoft-led open-source framework for building multi-agent conversations and orchestrated AI workflows.
AutoGen is used to create collaborative agent systems, tool-using workflows, and experiments where multiple LLM-driven roles coordinate on tasks.
Significant Gravitas
An open-source autonomous agent project and ecosystem for goal-driven AI task execution.
AutoGPT is used to experiment with autonomous workflows, recurring task execution, and broader agentic operating models over tools and data.
CrewAI
A framework for coordinating multiple role-based AI agents inside structured workflows.
CrewAI is used to build multi-agent systems where different specialized agents collaborate on planning, research, generation, or automation tasks.
LangGenius
A platform for building AI apps, workflows, and knowledge assistants with both hosted and self-hosted deployment patterns.
Dify is used to create internal copilots, RAG apps, and workflow-driven AI products without building every orchestration layer from scratch.
Stanford NLP
A framework for programming LLM pipelines with declarative modules, optimization, and evaluation-oriented workflows.
DSPy is used to build and tune prompt programs, retrieval pipelines, and structured multi-step AI systems with an emphasis on measurable performance.
Flowise
A visual builder for LLM workflows, agents, and chat systems with self-hosted deployment options.
Flowise is used to prototype chains and agents visually, connect models and tools, and deploy internal AI utilities without hand-writing every orchestration step.
deepset
An open-source framework for building search, retrieval, and question-answering pipelines for LLM applications.
Haystack is used to create RAG systems, search applications, and production AI pipelines that combine retrieval, ranking, and generation.
DataStax
A visual builder for composing LLM chains, tool flows, and AI applications with drag-and-drop orchestration.
Langflow is used to prototype retrieval workflows, wire tools and prompts together, and move from experimentation to packaged AI applications.
LlamaIndex
A framework focused on knowledge retrieval, data connectors, and agent workflows over structured and unstructured data.
LlamaIndex is used to connect data sources to models, build RAG systems, and create AI applications that need indexing and retrieval logic over business content.
Mastra
A TypeScript-focused framework for agents, workflows, and production-ready AI application building blocks.
Mastra is used to build agentic applications, tool-integrated workflows, and long-running automation in modern JavaScript and TypeScript stacks.
Microsoft
A managed studio for building enterprise copilots, agents, and connected workflow experiences on Microsoft's stack.
Copilot Studio is used to create internal assistants, connect enterprise data and actions, and publish governed copilots for business teams.
n8n
A workflow automation platform with strong support for AI agents, custom logic, and self-hosted deployment.
n8n is used to build automations that connect APIs, databases, and AI services, especially when teams want more control than a fully managed no-code platform offers.
OpenAI
OpenAI's framework for building agentic applications with tools, memory, and model-driven control loops.
The OpenAI Agents SDK is used to create tool-using assistants, orchestrate multi-step tasks, and structure agent application code around OpenAI models.
Pydantic
A typed Python agent framework centered on structured outputs, validation, and ergonomic developer tooling.
PydanticAI is used to build agent applications that need reliable schemas, tool interfaces, and stronger runtime validation around LLM interactions.
Microsoft
An SDK for building AI agents and orchestration workflows in .NET, Python, and Java.
Semantic Kernel is used to structure prompts, plugins, memory, and planning logic in applications that need reusable AI orchestration inside standard software projects.
SuperAGI
An agent platform for orchestrating autonomous workflows, tools, and business automation scenarios.
SuperAGI is used to build task-oriented agents, coordinate tool usage, and automate internal workflows around knowledge and actions.
Princeton NLP
An open-source research agent for software engineering tasks such as issue resolution and repository-level problem solving.
SWE-agent is used to experiment with agentic software engineering, benchmark task completion, and automate controlled coding workflows over repositories.