The source images for this workflow is attached as a zip file.<span style="background-color:rgb(26, 27, 30);color:rgb(141, 141, 143);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">This workflow requires my Comfyui-lmstudio-prompt-enhancer custom node which is installable via Comfy Manager.<a target="_blank" rel="ugc" href="https://github.com/AllenCraigBarnard/comfyui-lmstudio-prompt-enhancer">Learn more about this custom node</a>🛈 IMPORTANT!!!You will need to search for, download, and load the model google/gemma-4-12b-qat in LM studio. You will also need to start the API server and generate an API Key. <a target="_blank" rel="ugc" href="https://github.com/AllenCraigBarnard/comfyui-lmstudio-prompt-enhancer#lm-studio-setup">Learn how to configure LM Studio here.</a><h3 id="notes-for-using-googlegemma-4-12b-qat">Notes for using google/gemma-4-12b-qat</h3><ul><li>I chose the QAT variant over the non-QAT variant as this workflow can benefit from optimized memory utilization.</li><li>This is a memory efficient model which is fast and does a decent job at generating aesthetically pleasing prompts.</li><li>I set the max tokens to 2500, this higher value is to accommodate the reasoning tokens.</li><li>If the image looks flat, check the logs in LM Studio, the last log entry will contain a JSON path x.choices[0].message.content if the content field is empty that means the LLM ran out of reasoning tokens, this can be resolved by increasing max_tokens.</li></ul>

Files with SHA256: 66a0ba23d2b1257b50e7acce8d2a981113987241a32eaeae097ede541422fbfa