Release of Version 3 of the Photorealism ModelI am sharing the third test iteration of my photorealism model. This update introduces full manual labeling, giving you control over image modification through instructions.Key improvements in version 3Detailed control of instructions: All images in the dataset of 678 pairs have been manually labeled. This allows you to request specific changes or the preservation of specific features directly in the instructions.If the model insists on modifying things that weren't requested, reduce its intensity a bit or try to provide more detail in the prompt.Training steps: 1500<ul><li>Trigger Word: <code>transform to photorealism</code></li></ul>For example:- transform to photorealism, keep background unchanged- transform to photorealism, keep the fur skin- transform to photorealism, reduce head sizeSampler / Scheduler:- Euler / Simple- Euler / Linear_Quadratic &lt;- Strong modificationI'd love to see the tests you've done using the model. Feel free to post them.========================================What’s New in v2Dataset Expansion: Increased to 674 high-quality professional photos with broader contextual variety to enhance overall model comprehension.Lighting &amp; Effects: Added diverse lighting conditions and volumetric light effects for better depth and atmospheric accuracy.New contexts added:- Vector art- Modern anime- Digital painting- Manga- Cartoon- Oil paint- 8-bit pixel art- Simple draw sketchEverything is geared toward conversion specifically to photorealism while maintaining consistency, though it can also yield good results in other styles.It has not been tested or designed for the 6-step Viggle model, therefore I am not sure if it will have good results if used in conjunction with it.Training Steps: Extended from 250 to 1500 steps.<ul><li>Trigger Word: <code>transform to photorealism</code></li></ul>Sampler / Scheduler:- Euler / Simple- Euler / Linear_Quadratic &lt;- Strong modification==================================================<h3 id="first-test:-photorealism-model">First Test: Photorealism Model</h3>Sharing my first practical test in training a model focused on transforming images into photorealism. Technical Specifications<ul><li>Software: Ostris (ai-toolkit)</li><li>Dataset: 172 high-quality images (2K resolution), with all "before and after" pairs manually aligned.</li><li>Trigger Word: <code>transform to photorealism</code></li><li>Train Steps: 250 (currently; planning to extend training to 2,500–5,000 steps).</li></ul>MethodologyTo ensure that the final render quality is grounded in real-world details rather than synthetic approximations, I used a reverse workflow: real photographs were used as the AFTER state, which were then converted into vector art using the Qwen Image 2.1 model to serve as the BEFORE state. This guarantees that the learning baseline is anchored in genuine realism. Recommended Settings<ul><li>Sampler / Scheduler: Euler / Simple</li><li>CFG: 4</li><li>Steps: 25–40</li></ul><blockquote>Important Note regarding ComfyUI:The official text encoder currently has incorrect standard multiples. To achieve zero pixel drift and maintain maximum fidelity, it is recommended to use: <ul><li>Multiples of 28 for the VL</li><li>Multiples of 16 for the VAE</li></ul></blockquote>Since this is my initial test with a relatively low step count, I appreciate your understanding if some outputs are not yet fully refined. Feedback is always welcome!

Transform To photorealism

What’s New in v2Dataset Expansion: Increased to 674 high-quality images with broader contextual variety to enhance overall model comprehension.Lighting &amp; Effects: Added diverse lighting conditions and volumetric light effects for better depth and atmospheric accuracy.New contexts added:- Vector art- Modern anime- Digital painting- Manga- Cartoon- Oil paint- 8-bit pixel art- Simple draw sketchEverything is geared toward conversion specifically to photorealism while maintaining consistency, though it can also yield good results in other styles.It has not been tested or designed for the 6-step Viggle model, therefore I am not sure if it will have good results if used in conjunction with it.Training Steps: Extended from 250 to 1500 steps.Technical SpecificationsSoftware: Ostris (ai-toolkit)Trigger Word: transform to photorealismDataset Pipeline: All "before and after" pairs manually aligned (real photos as AFTER, Qwen Image 2.1 style conversions as BEFORE).- Since the AFTER state images consists exclusively of professional photos from real life like ocasions, Lora might try to tone down the extremely exaggerated proportions, especially the head size. (Try reducing the intensity a bit in these cases, if that is the goal.)Recommended SettingsSampler / Scheduler:- Euler / Simple- Euler / Linear_QuadraticCFG: 4Steps: 25–40Seed: AnyImportant Note regarding ComfyUI:To prevent pixel drift and preserve maximum quality, use:Multiples of 28 for the VLMultiples of 16 for the VAEFeedback and test results are always welcome!

Files with SHA256: f5548be5532854fcd725b38a0606ed2ceb6e924c57287fbd533c611b25fa3185