@Dr. Furkan Gözükara I'm having very blurry results with inpainting (I'm using the flux dev model no
@Dr. Furkan Gözükara I'm having very blurry results with inpainting (I'm using the flux dev model not the inpaint model maybe that's why?)
Fi have poco x6 pro. i gave my previous phone poco x3 pro to my wife :d
Fyou can make a decent one
Fbut wont be as good as local training
Fweird. but flux fill model best for inpainting
Awhat are the recommended parameters for inpainting?
FYouTubeSECourses
FLUX.1 Tools by BlackForestLabs changed the #AI field forever. They became the number 1 Open Source community provider after this massive release. In this tutorial, I will show you step by step how use FLUX.1 Fill model (inpainting model) to do perfect outpainting (yes this model used for outpainting) and inpainting. Moreover, I will show all fe...
4Yeah, that’s a shame, I would love a way to use those credits into a 128 rank Lora
4But at least I’m gonna test my dataset
4If it’s good enough
Wanybody tried the video gen in qwen2.5-max? i'm seriously impressed
Fqwen can generaste videos now?
![Dazzastrous [4090]](https://cdn.discordapp.com/avatars/871389071686135838/c4a5f4d81d9eeb18d856295d692e2c07.webp?size=40)
Dyep
Dcome on Doc your falling behind 
![Dazzastrous [4090]](https://cdn.discordapp.com/avatars/871389071686135838/c4a5f4d81d9eeb18d856295d692e2c07.webp?size=16)
Fye i have been really busy with real life 
Fcant access hugging face to check
Fi think it is not open source right?
Fi see api only
Wits not open source unfortunately, alibaba gatekeeping lol i've been using their interface https://chat.qwenlm.ai/
Qwen Chat offers comprehensive functionality spanning chatbot, image and video understanding, image generation, document processing, web search integration, tool utilization, and artifacts.
Wroughly 20-30min per 5 sec of video though so be prepared to wait
Fwell already so many such models and services
DThink its right up there if not better than kling
ADoes anyone know why is this glitch on the face happening? I'm using kling
A@Dr. Furkan Gözükara
AAPrompt : The man is doing pushups | Creativity Knob = 0.7
AOk lowering the knob kinda got rid of it but there's some still some left
N@Dr. Furkan Gözükara Does Torch 2.5 also provide speed boost for FLUX generation with SwarmUI? Or is it solely for Kohya training?
Fkling issue
Fye kling related
Fi think can be compared
Factually torch 2.6 published
Fi will test it
EHi @Dr. Furkan Gözükara. When you did the cogvideox tutorials, you must have done quite a few videos in the process, right?
A@Furkan Gözükara SECourses how can I have 2 characters of 2 different loras in the same photo? Only with inpaint / regional prompting?
Fi generated quite a few
Finpaint can ofc also work
Ahow do I do the regional? any video / document I can study?
Fi dont have any yet. look swarmui documentation
Fhe made some updates
Fi am trying to grab rtx 5090 no luck so far. some sells ultra expensive
Ahave you tried it tho yourself? did it work?
Fi had minimally tried before
Edid you eer do anything i2v fom illustrated images? Because I am getting nothing but distorted outputs. Trying to figure out if cogvideo even can do illustration animations
Fwell i did only the ones shown in tutorial video
FPatreon
Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon
DAn now then, is this the newly improved version of cogvid which has better coherence when it comes to following the prompts and better rendering? https://github.com/Vchitect/RepVideo
GitHub
The official implementation of ”RepVideo: Rethinking Cross-Layer Representation for Video Generation“ - Vchitect/RepVideo