AI Frameworks

Lightweight Python framework for building multi-modal agents. Memory, knowledge, and tool use built-in.

Microsoft's multi-agent framework. Agents collaborate via conversation to solve complex tasks autonomously.

Anthropic's TypeScript SDK for building Claude-powered agents. Tool use, multi-turn, and structured output.

Framework for orchestrating role-playing AI agents. Define crew roles, goals, and tasks for collaborative problem solvin...

Programming framework for LLMs by Stanford. Compile declarative programs into optimized prompts automatically.

Drag-and-drop UI for building LLM flows. No-code/low-code alternative to LangChain with visual pipeline editor.

Microsoft's constrained generation library. Interleave generation, prompting, and logical control for reliable output.

End-to-end NLP framework by deepset. Pipeline-based architecture for search, QA, and RAG applications.

Structured output extraction from LLMs using Pydantic. Retry logic and validation built-in. Python, TS, Ruby, Go.

Most popular framework for building LLM applications. Chains, agents, memory, and tool use with large ecosystem.

JavaScript/TypeScript port of LangChain. Full feature parity with Python version for Node.js and edge runtimes.

Visual framework for building multi-agent and RAG applications. DataStax-backed with cloud deployment option.

LangChain's framework for building stateful multi-agent applications with cycles, branching, and persistence.

Deploy LangChain runnables as REST APIs. Auto-generated docs, playground UI, and streaming support.

Framework for LLM agents with long-term memory. Virtual context management for unbounded conversation history.

Unified API for 100+ LLM providers. Drop-in OpenAI replacement with load balancing and spend tracking.

C/C++ LLM inference engine. Run quantized models on CPU/GPU with minimal dependencies. GGUF format standard.

Data framework for LLM applications. Specializes in RAG and connecting LLMs to external data sources.

TypeScript framework for AI agents and workflows. Built-in RAG, tool use, and workflow orchestration.

Memory layer for AI applications. Persistent, adaptive memory across sessions for personalized AI experiences.

Workflow automation platform with AI nodes. Connect LLMs to 400+ integrations. Self-hostable with fair-code license.

Official OpenAI framework for building multi-agent systems. Handoffs, guardrails, and tracing built-in.

Structured generation library. Guarantees valid JSON, regex-constrained output from any open LLM.

Agent framework by the Pydantic team. Type-safe agent definition with dependency injection and structured responses.

Microsoft's SDK for integrating AI into .NET, Python, and Java apps. Native Azure OpenAI and plugin system.

Hugging Face's lightweight agent library. Code-based agents that write and execute Python to solve tasks.

Spring ecosystem integration for AI. Portable API across providers with Spring Boot auto-configuration.

Hugging Face's production-ready inference server. Optimized for Transformer models with tensor parallelism support.

TypeScript toolkit for building AI apps. Streaming, tool calling, structured output with React/Next.js integration.

High-throughput LLM serving engine. PagedAttention for efficient memory, continuous batching, OpenAI-compatible API.