Dataset: AI Chat Models Performance in n8n Notification Workflows

This dataset contains structured performance logs of four open-source AI chat models (Minicpm-o-2_6, Gemma-3-12B-IT, Pixtral-12B, and Qwen2-VL-7B-Instruct) deployed locally through LM Studio and integrated with n8n, an open-source workflow automation tool. The dataset was created to evaluate the efficiency of model deployment in automated notification workflows, focusing on both inference performance and workflow orchestration overhead. The dataset provides detailed information on: - System metrics: CPU usage (%), GPU usage (%), and memory usage (MB) logged per second. - Performance metrics: Model response time, workflow latency, and success status. - Scenario metadata: Test ID, scenario name, input type (text-only, text+image, image-only), task purpose, and initiation method (Telegram or Webhook). Companion files are included: - XLSX/CSV data files with all logs. - README.md documenting dataset content and scenarios. - n8n workflow (JSON import file) for replicating the experimental workflow design. This dataset supports benchmarking, reproducibility studies, and comparative analysis of local AI deployments, with potential applications in AI efficiency research, multimodal systems, workflow automation, and sustainable computing.