Este repositorio ofrece la colección en formato GGUF del modelo original UltraReal Fine-Tune. Optimizados para inferencia con poca memoria.

Para obtener los mejores resultados con esta versión GGUF, se recomiendan los siguientes ajustes:

V4 Alright, so what’s new in this version? I cranked up the aesthetic dial, added more diversity in ages, and improved how it handles Asian features. But - because there’s always a but - I did notice the hands got a little wonkier. Eh, can’t win ‘em all.I highly recommend pairing this with my LoRAs, like the realism amplifier, 2000s analog core, and others, since this checkpoint works best as a base for stylized LoRAs. Might do one more version (because, let’s be real, I kinda scuffed v3 and v4 a bit), but first, I’m diving into fine-tuning Flex.Alpha. This time available versions: bf16, fp8, q8_0 - pruned fp16 name and q4_k_m - pruned fp8 name P.S: Don't use my UltraRealPhoto LoRA with this checkpoint - it has a huge impact on style, so image become overbaked. If you're using the UltraReal Fine-Tune, go with Realism Amplifier instead for the best results. UltraRealPhoto LoRa was created to fix crappy shadows, light and faces, but all that stuff already baked inside checkpoint, you can just add amplifier for better realism

V3 Update (Experimental) This release marks a step forward, although it’s still very much a work in progress. I focused on improving several key aspects, such as nudes, feet, and lower body anatomy. While the results are better than before, they’re not yet at the level I’m aiming for. That said, this version brings noticeable quality and texture enhancements, offering more detailed and refined outputs compared to the previous versions.

Recommended Settings:

CFG Scale: 3 (instead of 2.5 used in earlier versions)

Steps: 50 (helps with stability, though some minor instability remains in hands and fingers)

CFG 0.9 vs. 1.0: Lower CFG on 0.1 or even 0.2 may sometimes improve some details (may not improve, so feel free to expriment with this too), though it might take longer to generate.

Regarding nudes: they are still not working as intended, but I’m actively working on this issue and expect to address it in the next version.

The good news is that I already have the datasets prepared for V3.5, which I aim to release much faster than the gap between V2 and V3. With more experience and feedback from this version, I’m confident the next update will deliver significant improvements.

As always, I truly appreciate your support and feedback - it’s invaluable as I continue refining this project ❤️

P.S.: I feel like the more I fine-tune Flux, the more it degrades in other areas. Also i thought about trying finetune Flex Alpha (project looks very promising)

What's New in v2.0?

Enhanced Anatomy: Hands, feet, and poses have seen major improvements, offering more natural and accurate results. Say goodbye to overly distorted limbs!

Improved Textures & Quality: Upgraded skin details, richer textures, and sharper results overall. Blurred images still happen occasionally, but much less frequently than in the previous version or when using LoRAs alone.

Improved Text Rendering: Efforts have been made to improve the generation of text in images, and it’s much better than before. However, artifacts can still occur, and strange symbols might sometimes appear instead of readable words. This remains a work in progress.

Expanded Dataset: A larger and more diverse dataset (1800 images) introduces better balance across styles, lighting, and compositions.

Added Checkpoint VariationsTo ensure compatibility with different workflows, I’ve included multiple checkpoint variations:

BF16

FP8

Quant 8 (Q8)

Quant 4 (Q4) NF4

From my testing, I’ve noticed Quant 8 (Q8) offers slightly better quality than FP8, providing finer details while maintaining manageable resource requirements, but other works nice too. Pick the version that works best for your setup

Known LimitationsNSFW Capabilities: Still a weak area in this version. However, a minor fine-tune focusing specifically on NSFW content is already in the works.

Text Rendering: While text generation is better, occasional artifacts like odd symbols or incomplete words may still occur. But noticied usage of t5xxl fp16 instead of fp8 helps a lot with text

Tips for Optimal ResultsSampler: Use DPM++ 2M samplers for smooth and consistent outputs.

Steps: Aim for 30–50 steps to capture finer details without over-processing.

Scheduler: Beta Scheduler remains the best choice for this checkpoint.

Prompting Tips

The best prompting style involves complex prompts with clear, comma-separated phrases. While you can get creative with storytelling prompts, unnecessary descriptions like “this crap added more vintage to her style” won’t improve the results. Keep it concise and descriptive, focusing on essential visual details for the best output.

Future PlansI’m committed to further developing this fine-tune. The next update will likely focus on:

Expanding NSFW capabilities

Enhancing edge cases like dynamic poses and lighting scenarios

Improving text rendering for sharper, more accurate results

P.S: If you still don't have realistic effect, then try add my ultrareal lora, usually helps me a lot

Ultra-Realistic Flux Fine-Tune v1

This is my first experiment in fine-tuning a checkpoint, built upon the foundations of my UltraReal LoRA and expanded with an extended dataset. The aim? To push realism to the next level, finding that sweet spot between amateur aesthetics and professional, high-quality visuals.

While this is only the first version and I see room for further refinement - the results are good, but not ideal (hands and feet can be broken sometimes, but situation is not critical, still better then defaul flux). This fine-tune isn’t just about amateur-quality outputs; it shines with professional-grade images, offering exceptional detail, lifelike shadows, and lighting. It’s a versatile model designed to unlock a wider range of realistic image generation possibilities.

This is very much a work in progress, and I’m sharing it to gather feedback and see how others use it creatively. If you test it, I’d love to hear your thoughts or see your results! Also i uploaded both versions: fp16 (in ComfyUI it's better to use with e5m2) and fp8 and Q4_0

🌟 What’s New in This Fine-Tune?Expanded Dataset: Nearly double the dataset size of the original LoRA, covering a diverse range of styles, lighting, and compositions.

Improved Realism: Sharper details, richer textures, and more natural lighting, bridging the gap between AI-generated and real-world imagery.

Versatility: From casual amateur-style snapshots to cinematic, professional-quality renders, this fine-tune adapts to a variety of creative needs.

Enhanced Anatomy: Better hands, limbs, and more natural poses compared to the base Flux model.

💡 Tips for Best ResultsUse DPM++ 2M samplers for smooth and consistent outputs.

Aim for 30–50 steps for finer details without overdoing it.

Select the Beta Scheduler for optimal rendering performance.

⚡ Why Fine-Tune?This fine-tune was crafted to overcome some of the limitations of the default Flux model. It enhances its ability to handle complex scenes while maintaining consistent quality across a range of prompts. The goal is simple: make ultra-realistic image generation accessible, reliable, and visually stunning, without requiring endless adjustments.

P.S: i plan to train this model more to make ultimate checkpoint with best anatomy and realism. This version is not very good with nsfw (this will be fixed in next version) P.S.S: so far you can randomly get a low resolution image (dunno what exactly trigger this one, but will search for fixes). But seems like using high-resolution in prompt helps