🖥️Welcome to try out the open-source GPT4V-Image-Captioner, developed by my friend and me. It offers a one-click installation and comes integrated with multiple features including image pre-compression, image tagging, and tag statistics. Recently, we also launched the webui plugin version of this tool, everyone is welcome to use it!
🌍欢迎加入QQ群"兔狲·AIGC梦工北厂"，群号 ：780132897 ；"兔狲·AIGC梦工南厂"，群号 ：835297318（入群答案：兔狲）。Telegram群聊“兔狲的SDXL百老汇”，链接：https://t.me/+KkflmfLTAdwzMzI1
This model is a run-accelerated version of the HelloWorld SDXL base model, combining both SDXL Turbo and LCM technologies. Paired with the Eular a sampler, it can generate images within 6-8 steps, which is 3 times faster than the original SDXL version.This model is optimized for the Eular a sampler and it is recommended to only use the Eular a sampler for output.
After multiple rounds of testing, we identified the optimal integration ratio of the SDXL Turbo and SDXL LCM models. The current test results show that for the same 8-step image generation, the effect is: Turbo+LCM dual fusion > Turbo single fusion > LCM single fusion.
The image quality of the 8-step output from the Turbo+LCM dual fusion version is very close to the HelloWorld original model!
The memory usage of the Turbo+LCM dual fusion version is consistent with the HelloWorld original version. Therefore, if you have enough memory, it is recommended to enlarge the direct output image by 1.5 times (still within 6-8 steps).
The recommended parameters for generating images with this model are:
Sampler: Eular a (Important! The model is specifically adapted to Eular a, other samplers may not yield as good results)
CFG scale: 2 (Important! It is recommended to have a CFG scale between 1.5~2.5)
Sampling steps: 8 steps (6~8 steps are acceptable)
Hires algorithm: ESRGAN 4x (Other upscaling algorithms can also be used, not a mandatory option. Please ensure that your GPU memory is sufficient)
Hires Upscale factor: 1.5x
Hires steps: 8 steps
Hires Denoising strength: 0.3
本模型为HelloWorld SDXL原版结合SDXL Turbo和LCM技术的运行加速版本。搭配Eular a采样器，可以在6-8步内生图，是原sdxl版本的三倍速。本模型针对Eular a采样器进行效果调优，只推荐使用Eular A采样器出图。
最新版经多轮测试，得到了SDXL Turbo以及SDXL LCM两种模型的最佳融合比例，目前的测试结果是同样8步生图，效果上：Turbo+LCM双融合>Turbo单融合>LCM单融合。
Turbo+LCM双融合版本的8步出图画质已经非常接近HelloWorld原版模型！
Turbo+LCM双融合版本在内存占用上与HelloWorld原版一致，因此如果内存足够，建议对直出图进行1.5倍放大（同样6-8步），加速版模型可以用与xl原版大模型直出1024分辨率图像相近的时间，实现1024分辨率出图+1.5倍放大。
本模型推荐的生图参数：
采样器：Eular A（重要！模型针对Eular a专门适配，其他采样器效果不佳）
采样步数：8步（6~8步均可）
CFG scale：2（重要！CFG scale建议1.5~2.5）
放大算法：ESRGAN 4x（其他放大算法也可以，非必须选项，请确保GPU显存充足）
放大倍数：1.5倍
放大步数：8步
放大降噪系数：0.3
Description
FAQ
Comments (9)
Looking for just turbo and not turbo+lcm because Turbo has high quality potential but at least in my testing LCM can't achieve the same quality as base abut turbo seems to do even better with more steps.
不知道是不是我的web-ui参数问题，怎么感觉所有的Turbo或LCM模型，直出正面人脸还可以，但是张力大一点的动作图，就容易崩呢？
我最近切换到SDXL+Turbo+LCM的时候也遇到这个问题。找到解决的方法了么？
Thanks for the hard work! All I get is faces, upper body shots are very hard to get I don't know what I'm doing wrong or is a model mainly biased to faces?
If you set the cfg to 1 or 2 it generates horriffic screaming faces.
It's not the "smartest" model, but works quick and well.
Why not the smartest, example: it would put panties on the models even if the negative prompt said (panties:3.5) with no underwear related positive prompts. Otherwise very high quality output.
thanks for your work!
it comes a problem when i use this turbo model and adetailer, i get an unexpected square area above face (where adetailers works) in outcome picture which is not occured in another LEOSAM's HelloWorld model :(
I tried many methods attempt to fix it but failed, can i get any help? Thanks again~
谢谢你的工作！
我遇到了一个问题，当我使用这个turbo模型和adetailer时，我会遇到生成的人脸上出现一个灰色的正方形区域（adetailer的工作区域），在另一个LEOSAM's HelloWorld模型并不会发生:(
我尝试了很多方法试图修复他，但失败了。我能获得一些帮助吗？再次感谢~
Most of my results have typical upscaling artifacts adjacent to the eye-nose areas and on the above the lips (before upscaling), I don't know how LCM or Turbo works so it might be a problem inherent to one of those but it feels sus to me.
It also generates horrifying screaming faces if you use a really low cfg. I'm scared.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.