Home Creators Posts Import Register Favorites Logout

Content

Qwen Image base model vs FLUX SRPO training on the same 28-image dataset: 20 side-by-side results show Qwen far ahead at complex prompts and emotions.

Top images are Qwen and bottom ones FLUX. The difference is massive: Qwen destroys FLUX at complex prompts and emotions, so there is no return to FLUX. Links the Qwen Image training tutorial, the trainer app, the used prompts and full-size comparison images.

Full tutorial link > https://www.youtube.com/watch?v=DPX3eBTuO_Y

Qwen Image Models Training - 0 to Hero Level Tutorial - LoRA & Fine Tuning - Base & Edit Model

Info

This is a full comprehensive step-by-step tutorial for how to train Qwen Image models. This tutorial covers how to do LoRA training and full Fine-Tuning / DreamBooth training on Qwen Image models. It covers both the Qwen Image base model and the Qwen Image Edit Plus 2509 model. This tutorial is the product of 21 days of full R&D, costing over $800 in cloud services to find the best configurations for training. Furthermore, we have developed an amazing, ultra-easy-to-use Gradio app to use the legendary Kohya Musubi Tuner trainer with ease. You will be able to train locally on your Windows computer with GPUs with as little as 6 GB of VRAM for both LoRA and Fine-Tuning. Furthermore, I have shown how to train a character (person), a product (perfume) and a style (GTA5 artworks).

Resources

The post used in tutorial to download zip file : https://www.patreon.com/posts/qwen-trainer-app-137551634

Requirements tutorial : https://youtu.be/DrhUHnYfwC0

SwarmUI tutorial : https://youtu.be/c3gEoAyL2IE

Used Prompts

https://gist.github.com/FurkanGozukara/069523015d18a3e63d74c59257447f5b

Comparison Images Full Size

0

Related posts

All posts in this topic: Research, Comparisons and Experiments · New to SECourses? Start Here: Best and Most Popular SECourses Posts

Files

Comments

Diggy Dre

Oh wow! The selfie game is on point with this one! lol. It looks good!

cool1

Thanks. From a quick look it does seem that the qwen images are mostly following the prompt better. Though maybe some might look slightly more real or have slighty more detail in the flux than the qwen ones. They seem to be numbered different than in the prompt. eg. prompt 5 says "werewolf hunter..." but the werewolf shows on image 4.jpg when you open that in a new window. It would be good if there could be a .zip link where we can download all the comparison images in .jpg format if that could be done (since the ones that had .png format were about 20 MB each). Maybe it would be useful if comparison info could be shown as some sort of score too, where for each set of 2 images, a score was given to one or both based on how close they matched the prompt, then a total or average score could be shown for all the test images, so it would give a value for the ai model in general. eg. for some images I assume flux spro might be a bit better (eg. the image with the flying vehicles where one looks like painted shark teeth - the bottom one - flux - seemed more like just painted on shark teeth and the qwen one looked more like an actual shark shape. So maybe flux followed that bit of the prompt a bit better). Maybe there could be scores for prompt following and another score for realism of the 2 models, based on an average of the scores for each image.

SECourses

in every case Qwen wins. but for realism it is not there yet. currently i am working on improving realism with a realism lora. will share results hopefully

Richard Nagy

I love the flexibility (and the Apache2.0 license) of Qwen, but the colours and the realism are not there yet. I’m still generating with Flux.Krea. It’s still the best for me.

SECourses

for realism i am preparing a new preset. you will like it

Richard Nagy

Good man! Thanks everything!

SECourses

you are welcome sorry for delay. pay attention to tomorrow hopefully

chuu

Great comparison, Flux SRPO is much better in terms of realism and also in terms of the amateur aspect of the image! Is it possible to train a Lora on your Radio app with the Flux SRPO model, or is this not yet available? Thanks for everything!

SECourses

It is totally available here : https://youtu.be/nySGu12Y05k follow that tutorial. the links content configs are all updated. the downloader has flux srpo. we use same config

Ec Jep

BTW In my experience providing detailed captions for qwen during training is more helpful that just a single unique word when you want to generate very detailed long prompts. The single phrase tends to generate garbled images the more complex the prompts. For simple prompts they operate about the same.

Ec Jep

some aspects srpo are better than qwen but if you want incredible flexibility, realism and creativity, I prefer qwen. I am learning how to get better at qwen to get the high quality and when you can tease it out the realism is much better and srpo. I have trained lots of models with both and srpo produces crips images but skin texture and other details get smoothed out. Each progressive model has gotten better with these aspects. IMO I also do believe detailed captioning with qwen improves quality of image details

SECourses

on the contrary i tested this and single token works better. if not working better for you there can be 2 cases. first your dataset is not consistent. second you are not properly prompting after training.

Leonid Wolf

ERROR: flash_attn-2.8.2-cp310-cp310-win_amd64.whl is not a supported wheel on this platform. my Python 3.12.0 CUDA 12.1

SECourses

yes that is why. you need to have python 3.10.11 installed as well. it wont impact your system or other apps : https://youtu.be/DrhUHnYfwC0 follow this tutorial so you will run all apps + torch compile and it will be max speed