An open-source framework for developing LLM applications with instruction following, tool usage, planning, and memory capabilities.
A high-accuracy large language model (LLM) optimized for AI agent workloads, offering superior performance in multi-step reasoning and function calling at a reduced cost.