Qwen3‑Coder is an agentic code model from Alibaba’s Qwen3 family. Its flagship variant (480B MoE, 35B active) excels at agentic coding, tool use, and browser tasks with long context (256K–1M tokens). It’s open‑source under Apache‑2.0 and integrates with Qwen Code CLI and Qwen‑Agent workflows.

Qwen3‑Coder

qwen3-coder

Open‑source Mixture‑of‑Experts code model by Alibaba for agentic coding and tool use.

TEN Framework is an open-source modular system for building autonomous AI agents capable of reasoning, memory management, and tool usage. Designed to reflect real-world constraints, it incorporates a Thought-Emission-Norm (TEN) architecture, encouraging agents to think before acting. The framework supports open-ended planning, persistent memory storage, and tool augmentation. Developers can use TEN to experiment with agent autonomy and deploy agents that simulate complex decision-making workflows.

TEN Framework

ten-framework

An open-source framework for building autonomous AI agents that interact with the real world through tools and memory.

Trae Agent is an open-source, LLM-powered command-line assistant developed by ByteDance for software development workflows. It supports multi-LLM providers, file editing, bash execution, tool integration, trajectory logging, and a modular architecture—ideal for debugging, patching, documentation, and research.

Trae Agent

trae-agent

Open-source CLI LLM agent for automating software engineering tasks via natural-language commands.

AIWaves Agents (repo name “agents”) is an open‑source framework for autonomous language agents that supports planning, memory (short‑ and long‑term), tool usage, web navigation, multi‑agent communication, human‑agent interaction, symbolic control via SOPs, and symbolic learning with backprop‑style updates.

AIWaves Agents

aiwaves-agents

Open‑source adaptive LLM framework for building, training & deploying autonomous language agents.

Swarm is an open‑source, educational multi‑agent orchestration framework from OpenAI. It allows developers to define agents with instructions and tools, and supports handoffs—agent transfers via function calls—for dynamic workflows. Swarm is client‑side, stateless between runs, MIT‑licensed, and ideal for exploring multi‑agent patterns, though not intended for production.

OpenAI Swarm

openai-swarm

Open‑source experimental SDK for lightweight multi‑agent orchestration via instructions and handoffs.

AutoAgent is an open‑source, fully automated LLM agent framework from HKUDS that allows users to build, customize, and deploy multi-agent workflows using only natural language. It features an agent editor and workflow builder for zero-code design, a self‑managing vector DB, support for multiple LLMs, and strong performance on GAIA benchmark compared to research-grade agents.

AutoAgent

autoagent

Open-source zero-code LLM framework to create and deploy multi-agent workflows via natural language.

ScreenAgent is an open‑source Vision Language Model agent that interacts with real computer screens via screenshot observation and mouse/keyboard actions, following a planning‑execution‑reflection loop. It supports multi‑step GUI tasks, dataset collection, and achieves positioning accuracy comparable to GPT‑4V.

ScreenAgent

screenagent

Open‑source VLM agent to control computer GUIs via mouse/keyboard planning and execution.

AutoML-Agent is an open-source multi-agent framework that uses LLMs to automate the full ML pipeline—from data ingestion to model deployment. Designed for diverse data modalities, it decomposes tasks across specialized agents (e.g. prompt, data, model, operation), with retrieval‑augmented planning and multi-stage verification to ensure robust and deployable model solutions.

AutoML-Agent

automl-agent

Open-source multi-agent LLM framework for end-to-end automated machine learning pipelines.

AgentScope is an open‑source, developer‑centric multi‑agent platform for building, orchestrating, and deploying LLM‑powered agent systems. It features explicit message passing, zero‑code drag‑and‑drop UI, built‑in tool and MCP support, distributed execution, monitoring, automatic prompt tuning, and fault tolerance.

AgentScope

agentscope

Open‑source multi‑agent framework with drag‑and‑drop UI, distributed execution, tool calls, and fault tolerance.

Strands Agents is an open‑source SDK from AWS for building production‑ready AI agents using a model‑driven architecture. It supports any LLM provider, rich tool integration (including MCP and AWS services), multi‑agent workflows, observability via OpenTelemetry, and deployment in local or cloud environments.

Strands Agents

strands-agents

Open‑source SDK for building and orchestrating single or multi‑agent systems with LLMs and tool integration.

MCP‑Use is an open‑source library and platform that lets developers connect any LangChain‑compatible LLM to Model Context Protocol (MCP) servers. It enables creation of AI agents with tool access—like browsing, file operations, or HTTP services—supporting multi‑server setups, dynamic tool selection, streaming output, and both hosted and self‑hosted deployment options.

MCP‑Use

mcp-use

Open‑source platform to connect LLMs with MCP servers and build custom AI agents with tool access.

Keploy is an open‑source AI test automation agent that uses eBPF to record live traffic and generate deterministic unit, integration, and API tests with mocks/stubs. It integrates with CI/CD, supports multiple languages, and uses multi‑model LLMs for test generation and self‑healing.

Keploy

keploy

An open‑source AI agent that auto‑generates and maintains unit, integration, and API tests with mocks.

Explore AI Agents with free pricing