Zleap-Agent

Automated alignment adjustment for LLMs — direct steering, LoRA, and MoE expert-granular abliteration, optimized via multi-objective Optuna TPE.

Agent-lightning is designed to enhance the capabilities of AI agents, enabling them to learn and adapt more effectively. By utilizing advanced techniques in machine learning and reinforcement learning, it provides a framework that simplifies the training process for building intelligent systems.

Give agents an operating system as a library. Runs in your existing backend – no sandboxes, VMs, or SaaS. Powered by WebAssembly & V8 isolates.

Give your AI agent eyes to see the entire internet. Useful for agent-infrastructure and ai-agent use cases.

Enhance your virtual agents with pre-built skills that allow them to interact more effectively and intelligently in various scenarios. This resource provides tools and examples to streamline the development process for creating smarter, more capable agents.

AgentCPM provides a comprehensive framework for training and assessing different language model agents. It simplifies the process of developing and evaluating these agents, making it easier for users to create effective AI-driven solutions.