Software Engineering Courses (SECourses)•10mo ago

Maybe I need to try again with later code. Also this guy just created a gui for it. https://github

Maybe I need to try again with later code. Also this guy just created a gui for it. https://github.com/Kvento/musubi-tuner-wan-gui there is also a gui from ttplanet https://github.com/TTPlanetPig/Gui_for_musubi-tuner - this is gradio

TtheJindesan Maybe I need to try again with later code. Also this guy just created a gui for...

Furkan Gözükara SECourses•3/17/25, 11:08 PM

i plan to make my own gradio

FFurkan Gözükara SECourses i plan to make my own gradio

theJindesanOP•3/17/25, 11:12 PM

I just modified TTplanet's one. its a good starting point. Let me know how your Wan Lora success is. I'm going to keep experimenting, and will let you know what i discover too. I think there is a reason character lora's aren't being released as fast as "motion/animation" loras

Furkan Gözükara SECourses•3/17/25, 11:13 PM

sure. i am testing 0.2 teacache right now

Furkan Gözükara SECourses•3/17/25, 11:13 PM

it takes less than 10 min on h100 highest quality - 20 steps

theJindesanOP•3/17/25, 11:15 PM

Also TorchCompileModelWan (for wan21) breaks Lora's on 50xx but not 30xx. Must be the 3.3 triton, or torch nightlies. I just tested. Same workflow works fine on 30xx. but not 50xx with Torchcompile running on both.

TtheJindesan Also TorchCompileModelWan (for wan21) breaks Lora's on 50xx but not 30xx. Must...

Furkan Gözükara SECourses•3/17/25, 11:15 PM

very likely

Furkan Gözükara SECourses•3/17/25, 11:16 PM

wow teacache 0.2 not bad

Furkan Gözükara SECourses•3/17/25, 11:16 PM

and super fast

theJindesanOP•3/17/25, 11:16 PM

Its hard to find people with direct 5090 experience still. Shortages are tough.

theJindesanOP•3/17/25, 11:17 PM

I was using 0.16 then 0.2 - but i was reading another thread where they reported 0.25 being the perfect spot

theJindesanOP•3/17/25, 11:17 PM

Not my research - direct from Kijai: Benchmarks. Everything here is Wan2.1 720 @ 49 frames, sageattention, cuda 12.6, 3080 ti, listing thresh, starts, times, and quality observations

0.300 (rel_l1_thresh) a lot more subtle grainy noise than 0.250
0.00 (start) -> 13:08 (encoding time)
0.2875
0.00 -> 13:39 more grainy noise
0.280
0.00 -> 13:18 92% starts noisy
0.275
0.00 -> 14:55, near indistinguishable from 0.25
0.270
0.00 -> 14:41, 96%, nearly perfect
0.265
0.00 -> 16:09, 96%, near perfect
0.250 -> looks lossless
0.10 -> 16:30

2x run:
Best so far, but slower,
0.250
0.10 (start) -> 16:20

Near if not lossless at 0.250, at the cost of about 3 minutes when compared to 0.300.

in summary

0.300 ~ 13:00 I found a bit too grainy for my liking, though speed is nice
0.275 ~ 14:55 great balance between speed and quality
0.250 ~ 16:30 I found was basically lossless

TtheJindesan Click to see attachment

Weatherby•3/18/25, 12:36 AM

That is odd...

Wolfparty7•3/18/25, 4:35 PM

Hello everyone

Wolfparty7•3/18/25, 4:35 PM

Can somone hel me to solve this?

Wolfparty7•3/18/25, 4:35 PM

Im trying to merge two models in SD

Wolfparty7•3/18/25, 4:35 PM

but im getting this error

Wolfparty7•3/18/25, 4:36 PM

Wolfparty7•3/18/25, 4:36 PM

Its a VRam problem?

art-ai_usa-girl•3/18/25, 8:48 PM

stable-virtual-camera https://stability.ai/news/introducing-stable-virtual-camera-multi-view-video-generation-with-3d-camera-control?utm_source=x&utm_medium=social&utm_campaign=virtual+camera

Stability AI

Introducing Stable Virtual Camera: Multi-View Video Generation with...

Introducing Stable Virtual Camera, currently in research preview. This multi-view diffusion model transforms 2D images into immersive 3D videos with realistic depth and perspective—without complex reconstruction or scene-specific optimization.

art-ai_usa-girl•3/18/25, 8:48 PM

model https://huggingface.co/stabilityai/stable-virtual-camera/tree/main

stabilityai/stable-virtual-camera at main

art-ai_usa-girl•3/18/25, 8:53 PM

https://github.com/Stability-AI/stable-virtual-camera?tab=readme-ov-file

GitHub

GitHub - Stability-AI/stable-virtual-camera: Stable Virtual Camera:...

Stable Virtual Camera: Generative View Synthesis with Diffusion Models - Stability-AI/stable-virtual-camera

WWolfparty7 Click to see attachment

Furkan Gözükara SECourses•3/18/25, 9:23 PM

yep. select cpu to load

Inoshi•3/19/25, 8:35 PM

with the wan 2.1 zip file you provide is triton also included or do i need to install it manually? also is teacache using more vram?

IInoshi with the wan 2.1 zip file you provide is triton also included or do i need to in...

Furkan Gözükara SECourses•3/19/25, 9:15 PM

included

Furkan Gözükara SECourses•3/19/25, 9:15 PM

teacache shouldnt impact vram

theJindesanOP•3/19/25, 9:34 PM

@Furkan Gözükara SECourses have you seen a xformers fix for blackwell yet? I searched but haven't seen progress. I saw someone compiled wheels for python 3.12 but not 3.10, was thinking to upgrade venv for kohya to 312. Also for wan21. lora training, you must use captions, or the lora model can't do anything, no flexibility beyond the source images similarity. I tried at least 4 tests using musubi. Also still struggling to get good likeness as compared to hunyuan

TtheJindesan <@205854764540362752> have you seen a xformers fix for blackwell yet? I searche...

Furkan Gözükara SECourses•3/19/25, 9:42 PM

i am working on this for weeks

Furkan Gözükara SECourses•3/19/25, 9:42 PM

https://github.com/facebookresearch/xformers/issues/1234

GitHub

xFormers is still broken for Torch 2.7 and CUDA 12.8 - we need it f...

Just compiled with below pip freeze (venv310) C:\f\x10>pip freeze annotated-types==0.7.0 colorama==0.4.6 deepspeed @ https://files.pythonhosted.org/packages/3d/65/1a6394f5d6dee851e9ea59e385f6d64...

Furkan Gözükara SECourses•3/19/25, 9:43 PM

https://github.com/facebookresearch/xformers/issues/1229

GitHub

xFormers for CUDA 12.8 : AttributeError: Cannot set attribute 'src'...

I have compiled xFormers on xformers-0.0.30+c5841688.d20250306 torch==2.7.0.dev20250228+cu128 triton-3.2.0+git8f9b005b the compile worked I am able to install. Python 3.10.11 - Windows 11 However I...

theJindesanOP•3/19/25, 9:43 PM

https://huggingface.co/Panchovix/xformers-windows-blackwell2.0-nightly/tree/main

Panchovix/xformers-windows-blackwell2.0-nightly at main

theJindesanOP•3/19/25, 9:43 PM

i can't get it to compile, but someone did.

TtheJindesan https://huggingface.co/Panchovix/xformers-windows-blackwell2.0-nightly/tree/main

Furkan Gözükara SECourses•3/19/25, 9:50 PM

i compiled many times

Furkan Gözükara SECourses•3/19/25, 9:50 PM

it doesnt work

cilium•3/19/25, 10:39 PM

Hi guys, do you know of any AI for subtle 2D animations that are clean?

cilium•3/19/25, 10:40 PM

Just very subtle movement for a cutout 2D image

Ccilium Just very subtle movement for a cutout 2D image

Furkan Gözükara SECourses•3/19/25, 10:43 PM

wan 2.1 is best atm

Hiramos_•3/20/25, 4:15 AM

@Furkan Gözükara SECourses Have you seen this? https://riflex-video.github.io/

It looks like an amazing new way for video extension

Iamme•3/20/25, 6:17 AM

Hey @Furkan Gözükara SECourses I am running the wan2.1 on H100 14b model, while text to video works great, I'm having issues with Image to video, I'm following every step of your tutorials but when I try to generate i get an error (no lens). what am I doing wrong?