Vikhrmodels/Qwen2.5-7B-Instruct-Tool-Planning-v0.1

Vikhrmodels/Qwen2.5-7B-Instruct-Tool-Planning-v0.1 is an experimental supervised fine-tuned version of Qwen/Qwen2.5-7B-Instruct, developed by Dmitry Tishencko and the Vikhr Team. This model specializes in advanced tool planning and function calling tasks, including handling simple, multiple, and parallel function calls, tool relevance detection, and robust error handling. It is specifically trained on synthetic English data from the Vikhrmodels/tool-plannings-v0.1 dataset, focusing on the model's ability to reason about tool usage before execution.