xingyaoww/CodeActAgent-Llama-2-7b
TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Jan 5, 2024License:llama2Architecture:Transformer Open Weights Cold

CodeActAgent-Llama-2-7b by Xingyao Wang is a 7 billion parameter Llama-2 based language model with a 4096 token context window, fine-tuned using the CodeActInstruct dataset. This model specializes in enabling LLM agents to perform executable Python code actions, integrating with a Python interpreter for dynamic action revision and new action emission based on observations. It excels in out-of-domain agent tasks by consolidating actions into a unified CodeAct space, demonstrating improved success rates over text and JSON action formats.

Loading preview...