# Gemini "Analyze an Image" node gets stuck on loading and times out **URL:** \ **Category:** Questions\ **Tags:** workflow-building\ **Created:** [May 21, 2026, 9:39am UTC](https://community.n8n.io/t/gemini-analyze-an-image-node-gets-stuck-on-loading-and-times-out/296423 "2026-05-21T09:39:58Z")\ **Posts on this page:** 1\ **Showing post:** 7
**Author:** ![tamy.santos](https://dub1.discourse-cdn.com/n8n/user_avatar/community.n8n.io/tamy.santos/32/153776_2.png) [@tamy.santos](https://community.n8n.io/u/tamy.santos)\ **Post date:** [May 21, 2026, 12:47pm UTC](https://community.n8n.io/t/gemini-analyze-an-image-node-gets-stuck-on-loading-and-times-out/296423/7 "2026-05-21T12:47:16Z")
@Product > **[Step 3: Computing metrics - Building your own LLM evaluation framework](https://blog.n8n.io/llm-evaluation-framework/#step-3-computing-metrics)** > > If you’ve ever built an application powered by Generative AI, you know the feeling: one small change to a prompt, a model swap, or a slight tweak to a node can turn a perfectly functional workflow into an unpredictable mess. Unlike deterministic... > [@Can one train AI for n8n?](https://community.n8n.io/t/189033): > > Hello, I’m trying to build an AI image analysis tool that writes an estimate, or at least identifies the damaged parts, of a car from a POST request with the VIN and an image of the damage. I vibecoded a site lol that sends JSON of the VIN and binary of the image via POST and then edits the image to put the VIN in the top left corner, and I got big company LLMs such as claude and chatGPT to read the vin in the top left then identify the damage and respond with a JSON format of the damage and the… ![image](https://europe1.discourse-cdn.com/n8n/original/3X/2/f/2fba832bc94fd5bd4b5f82fce384775470a6acea.png) --- _[View the full topic](https://community.n8n.io/t/gemini-analyze-an-image-node-gets-stuck-on-loading-and-times-out/296423)._