Mad Dream of AI

As open-source large language models continue to evolve, I attempted to utilize the Ollama platform by invoking the LLaVA model to interpret content across multiple images. Following this, I leveraged the Qwen2 model to structure these insights into narratives and prompts, ultimately feeding them into SD for creation of random outputs.
This endeavor was solely driven by the joy of exploration and curiosity.

[@Simon19770707]

這個流程的效果蠻有趣的，請問可以分享來學習參考嗎?謝謝

[@kevinhu1136]

噢。这个流程也是突发奇想，做了玩的，所以还没整理过。