Orchestration Frameworks

Orchestrates multi-agent systems built on the OpenAI Assistants API with role-based collaboration.

Evolves agent skills, prompts, and harness modules by merging diffs from parallel workers against held-out reward.

Alibaba multi-agent framework with distributed deployment and fault tolerance for production use.

Multi-agent framework with a runtime and control plane for managing agent deployments at scale.

Official Claude SDK with tool use, computer control, and streaming for building Claude-native agents.

Event-driven multi-agent framework merged with Semantic Kernel for production workflows.

Managed AWS infrastructure for agents with Bedrock models, enterprise compliance, and auto-scaling deployment.

Role-based multi-agent simulation framework for collaborative reasoning and task completion.

Open-source platform for embedding custom AI copilots into React applications.

Production multi-agent framework with 5K+ GitHub stars, role-based collaboration, async execution, and 1500+ company adoptions.

Stanford framework that programmatically optimizes LLM prompts using machine learning.

Self-hosted MCP-native agent platform with visual workflow automation and multi-provider support.

Modular agent dev kit with native Gemini and Vertex AI integration.

Production-ready AI orchestration framework focused on building customizable LLM applications and RAG pipelines.

Stateful, graph-based agent workflow framework from LangChain.

Stateful agents with built-in long-term memory and a REST API server.

Lightweight agent framework with memory, MCP/SSE, Tree-of-Thought planning, and LightSwarm collaboration.

The framework for connecting LLMs to your data, with indexing and retrieval capabilities.

No-code agent orchestration platform with visual workflow builder and enterprise deployment.

Opinionated TypeScript framework with RAG, observability, and MCP support built in.

Lightweight agent framework with self-editing prompts and code for minimal-overhead agent loops.

Serverless framework for high-throughput agent workloads with minimal cold starts.

Open-source phone agent framework for building mobile device automation agents.

Lightweight multi-agent SDK with tracing and guardrails from OpenAI.

Self-programming framework whose agents create, run, and refine workflows across models, tools, memory, and context.

Production multi-agent framework with self-reflection, MCP integration, and workflow automation.

Type-safe agent framework from the Pydantic team with a FastAPI-style developer experience.

Microsoft enterprise SDK for Python, C#, and Java with modular plugins, memory, and goal planning.

Hugging Face code-first framework where agents write and execute Python instead of JSON tool calls.

AWS model-driven agent SDK with native Bedrock integration.

Minimalist agent framework with MCP support and rapid setup for quick prototyping.