An open-source framework for building, evaluating, and training general multi-agent systems. Features natural language agent creation, distributed reinforcement learning training pipeline, and complex environment interactions. Ranks top on authoritative benchmarks including GAIA, OSWorld, and VisualWebArena.
An open-source framework by Stream for building vision AI agents that work with any model or video provider, leveraging Stream's edge network for ultra-low latency video experiences.
A Windows desktop application integrating chat, text-to-speech, image generation from text, and machine translation. It supports mainstream AI services and MCP server plugins, with full AOT compilation to deliver an excellent desktop AI experience.
Odyssey is a framework that empowers LLM-based Minecraft agents with open-world skills, featuring 40 primitive skills and 183 compositional skills, enabling AI to autonomously explore, learn, and execute diverse tasks in the Minecraft universe.
AdalFlow is a PyTorch-like library for building and auto-optimizing any LLM workflows, from chatbots and RAG to agents, featuring auto-differentiation for prompt optimization and model-agnostic building blocks.
An open-world multi-task agent system with memory-augmented multimodal language models that can understand visual observations and human instructions, generate sophisticated plans, and perform embodied control in the Minecraft universe, capable of completing over 200 diverse tasks with human-like control and observation.
MedResearcher-R1 is a comprehensive training data generation and synthesis framework for medical scenarios, built on a knowledge-informed trajectory synthesis approach that provides an end-to-end solution from knowledge extraction to model training data generation and evaluation.
An end-to-end reinforcement learning training framework for orchestrating tools and agentic workflows that coordinates various models and tools to achieve more efficient solutions than large language models alone.
An end-to-end, self-evolving training framework that unifies self-questioning, self-navigating, and self-attributing into a cohesive system, empowering agents to autonomously improve their capabilities for efficient, cost-effective, and continuous capability evolution.
ReCall is a novel reinforcement learning framework that trains LLMs to reason with tool calls, enabling them to solve complex problems without requiring supervised data on tool use trajectories or reasoning steps.
Page 1 / 1 · 10 total
Get the latest AI tools and trends delivered straight to your inbox. No spam, just intelligence.
