Home
Creators
Search Recent Random
Posts
Search DMs Popular Hash Lookup Tags Random
Importer
Import FAQ
Register Favorites Keys Review DMs Logout
Community
Telegram ThePornDude
Home Creators Posts Import Register Favorites Logout

1-Click INSTALL Ai-TOOLKIT WAN LORA TRAINING! (Patreon)

Published:
2025-07-20 16:50:43
Edited:
2025-07-20 16:51:07
Imported:
2026-06
Tags:

Downloads

Content

Hey everyone! I've created a 1-click installer for the “ULTIMATE WAN LORA TRAINING AI‑TOOLKIT WEBUI.” Hands‑down the BEST AI model training I have ever made! VIDEOS & IMAGES!!! Just fricking insane! This really is the best Ai model currently available and Ai-Toolkit makes training a Lora super ultra easy as well, basically the ultimate combination, perfect for complete beginners! With it you can train any LoRA you want for WAN: people, styles, and even full video LoRAs… it’s just nuts!

You can check out the video right here: https://youtu.be/lRg5sPBXTZEThe installer of course automates the entire install process!

LINK TO THE WORKFLOW + LORAS: https://www.patreon.com/posts/134551107LOCAL INSTALL!!! RECOMMENDED TO HAVE AT LEAST 24GB OF VRAM FOR THE WAN TEXT TO IMAGE LORA TRAINING! if it's an issue, definitely use the Runpod installer instead as recommended in the video!

Local install:

Requisites:

– CUDA 12.8 (RTX-50) OR CUDA 12.6 (RTX-40/30/20/GTX)

– Git, Python 3.10 64-bit, Node 18/20 LTS

1. Download the AI-TOOLKIT_AUTO_INSTALL.bat and place it in a folder WITHOUT any spaces in the path!!!

2. Run the bat file and select either option 1 or 2 depending on your GPU

3. Wait for the install to finish

4. ???

5. Profit 😎

Use then one of the launcher once you have closed the ui and you want to restart it.

RUNPOD:

1. Create an account if you haven't already: Runpod

2. Click on Pod (on the left side) then click deploy

3. Choose a 5090 GPU (if you intend on training a text to image lora OR a text to video lora with text encoder unloaded and 512 resolutions) otherwise choose the 6000 pro blackwell GPU, and chose the pytorch 2.8.0 then edit the template and choose 100GB for both the container and volume disk, then under expose ports, edit:

8888 into 8888, 8675. Then go into environment variables below, click add and then type:

AI_TOOLKIT_AUTH on the left then type a password on the right

then scroll then and deploy.

4. Go to my pods, wait for everything to finish and then click "connect", then "Connect to HTTP SERVICE port 8888"

5. Then drag and drop the AI-TOOLKIT_AUTO_INSTALL-RUNPOD.sh file on the left side of the UI then click on the "Terminal" icon on the right side on the UI

6. Copy and paste these two lines then press enter:

chmod +x AI-TOOLKIT_AUTO_INSTALL-RUNPOD.sh

./AI-TOOLKIT_AUTO_INSTALL-RUNPOD.sh

7. Wait for everything to be installed

8. Go back to the port 8675 and click on it to launch the webui then enter your password you created previously

IF YOU WANT TO RESTART THE UI AFTER STOPPING AND RESTARTING THE POD

Use the same

chmod +x AI-TOOLKIT_AUTO_INSTALL-RUNPOD.sh

./AI-TOOLKIT_AUTO_INSTALL-RUNPOD.sh

command as you did for the install

Chatgpt prompt:

I will upload a few images, wait until I finished uploading them all.

These are photos of Margot Robbie, analyze those images and caption them correctly for a Lora training using "Margot Robbie" as the caption token and convert each single caption into a .txt file and name them all in the same order, Margot Robbie_0001.txt, Margot Robbie_0002.txt, etc... and then finally zip them up all into a single zip file I can download

portrait of margot robbie, clean studio headshot, 85 mm lens, soft beauty lighting, photorealistic

margot robbie, film‑noir portrait, dramatic hard key light, deep shadows, black‑and‑white photo

oil painting portrait of margot robbie in the style of John Singer Sargent, loose brushwork, warm palette

margot robbie cyberpunk portrait, neon rim‑lights, rain‑soaked city backdrop, vivid colors, photoreal

1940s Hollywood glamour portrait of margot robbie, sepia tone, classic butterfly lighting, vintage lens flare

pop‑art portrait of margot robbie, bold halftone dots, saturated primary colors, Andy Warhol style

ukiyo‑e woodblock portrait of margot robbie, flat colors, delicate linework, Edo‑period aesthetic

renaissance oil portrait of margot robbie, chiaroscuro lighting, rich velvet dress, baroque background

futuristic holographic portrait of margot robbie, volumetric light, translucent interface elements, sci‑fi realism

candid Polaroid portrait of margot robbie, pastel color shift, soft film grain, natural window light

As always, supporting me on Patreon allows me to keep creating helpful resources like this for the community. Thank you for your support - now go have some fun😉!

Files

Comments

Marcus LeVere (edited)

Comment edits

2026-09-08 20:00 The A6000 Ada Lovelace is also option 1 on your GPU selection right?
2025-07-20 19:25

Aitrepreneur

no that's 2, number 1 is blackwell architecture

Manic

No module named 'dotenv' is preventing me from starting my lora training. help

frstgear

I am also getting a similar error: Traceback (most recent call last): File "C:\lora\ai-toolkit\run.py", line 6, in from dotenv import load_dotenv ModuleNotFoundError: No module named 'dotenv'

joe

Can you please tell me the settings if I want to train wan text to video on only images (not videos like you shown in the tutorial)

Aitrepreneur

Come on bro......... watch the video? Pretty please with cherries on top?

Yu Yeh

Appreciate this tutorial very much. On the text to video LoRa training, will it make a difference in quality if the short training clips' resolution are higher than 512 or 768 if you have set the training resolution parameter of the dataset to 512?

Shawn Saw

Where do I send error messages from the console?

Aitrepreneur

The dataset always needs to be as high quality as possible it doesn't matter at which resolution you train as you as your base dataset is high resolution and high quality

Thomas Collard

When i launch "Launcher toolkit" it just open and instantly close. what did i do wrong? I already tried to reinstall it multiple times but the issue remains. And i made sure there was no space in the folder name.

Manic

Yes. I did get it without with the help of Aitrepreneur

puk

I have the same problem on Windows. When i start the app it close instantly

Erik Simonetti

I am getting Cuda out of memory on a 3090 (24gb). Any idea if there is some tweaks that would make it fit? I have Unload TE enabled, like you say in the video. Thanks a lot!

Phil

turn off avg

Rob113

Any update on this? I am seeing the missing dotenv error also.

joe

I watched the video 2 times before I asked here, you mention only text to image with image training data and text to video with video training data, you don't say anything about text to video with images as training data

Aitrepreneur

open a cmd window and drag and drop the launcher inside then press enter, this will give at least an error we can use to troubleshoot why it happens. Dm me the info though

Aitrepreneur

open a cmd window and drag and drop the launcher inside then press enter, this will give at least an error we can use to troubleshoot why it happens. Dm me the info though

Aitrepreneur

yup what Phil said, turn off the AVG, it's a false positive (also tbh not sure why you got AVG, it's terrible...)

Wilhelm Zabłotnikov

If I train a lora for, say, all 3000 steps I planned and finish, can I then continue to train it further? Or do I have to start from the beginning? By the way, thanks for another excellent tutorial.

Aitrepreneur

you can train further, just click on the modify icon, add more steps and start

Aitrepreneur

But....how can you make a training for videos...solely with images?? The training is gonna go through if that's what your are asking but.. the while point of video generation training is that you are trying to copy the effect or look and motion of the dataset videos...

D Dunn

What is your method of choice for prompting your training images for Wan Loras?

Aitrepreneur

I just ask chatgpt to create me a "flux prompt" with the details i want

joe

Ok but how do people make character loras?

Erik Simonetti

Sorry I forgot to specify, I am trying to train a person lora on Wan 2.1 14B, using the same settings as you show in the video. I also trained a Flux kontext lora, following your previous tutorial and that worked fine.

Rob113

Thanks Aitrepreneur, That seems to have done the trick. It's chugging away happily now. :-)

Nicholas Nelson (edited)

Comment edits

2026-09-08 20:00 Cool stuff! Any chance we will see an updated chatbot install anytime soon?
2025-07-22 00:23

Samuel

3090 not eough vram even with 512 resolution. Im trying cahe text encoder and unload text encoder too..

LW

Hi K, thanks for the video and the one-click installer for WAN lora training. You previously did a one-click installer for Kontext which also use the AI-toolkit. Should I install the WAN lora one-click installer separately or are they essentially the same, both using the AI toolkit?

HEATT

Python was not found; run without arguments to install from the Microsoft Store, or disable this shortcut from Settings > Apps > Advanced app settings > App execution aliases. C:\AITOOLKIT\ai-toolkit>

HEATT

not all that good at this stuff, i downloaded and ran the python installer, setup my data set and launched it through the installer. Whenever i try to run job i get this. Also whenever i run the launch file bat in the ai toolkit folder it says please run installer first, AITI yes, but please help idiot

Aitrepreneur

yes, it's something I want to do, when? I don't know

Aitrepreneur

Install the Wan one, I've removed a few lines that were problematic for some people before

Glen

Thank you K its all working on Runpod 😁👍👍

Aitrepreneur

Make sure that you have the latest nvidia drivers and that you have the sysmem fallback option enabled in the nvidia control panel

Aitrepreneur

??? Again...watch the video.....what do you think margot robbie lora training is?

LW

Thanks K

robin poirieux

I try the clean bat but it not works on downloading orginal wan model. (https://github.com/comfyanonymous/ComfyUI/releases/download/v0.3.44/ComfyUI_windows_portable_nvidia.7z) The error is : " curl: (18) end of response with 508426747 bytes missing Download failed." I try with de command line: curl -L --max-time 0 --retry 5 --retry-delay 5 --continue-at - -o ComfyUI.7z "%COMFY_RELEASE%" but it not works too even when trying a manual download through the browser... a corrupt file from server ? busy server ?

Samuel

images into text2video model. In musubi in able, in AIToolkit using captions txt and 512 res is not letting me

Yu Yeh

Hello AI Lord! Do you require to pay for Network Storage over at Runpod if you want to save your ComfyUI's workflows and custom LoRas for on-going use? Thanks!

D Dunn

Does anyone know how (what file) to change the default sample prompts?

Aitrepreneur

I think it's the ai-toolkit/ui/src/app/jobs/new/jobConfig.ts

Vato Vegas

Having issues with the one-click installer for Ai-Toolkit all other one click installers work. When it finishes intstalling the window automatically shuts - when I run the LAUNCHER-TOOLKIT.bat manually it opens then quickly closes When I try to run as admin. it says "[ERROR] venv not found. Run the installer first." I am in a death loop the install has NO spaces can it handle under_scores?

Aitrepreneur

I don't personally use it myself but I'm guessing so, you always pay for stockage

Aitrepreneur

Are you using a VPN? This is a connection issue

Bob Winberry

Thanks for your videos and your work on explaining . I have however, not been able to get this last Lora Make to run as I get an error: Traceback (most recent call last): File "C:\AITOOLKIT\ai-toolkit\run.py", line 6, in from dotenv import load_dotenv ModuleNotFoundError: No module named 'dotenv' I've tryed ingalling it, but says it's already installed. I deleted it and installed again, but no luck. Any ideas? Thanks again

J.ruthless

Hey im actually having the exact same issue. All is installed properly as per your instructions and I have installed all other one clicks with no issues. But the launcher closes immediately after I try to run it. Saying venv not found

J.ruthless

Hi I sent you a dm aitrepreneur

Timmy

I am also running into this issue. Can I DM you also?

Bob Winberry

I just tried that, no luck, so I'll just move on from this one - thanks for your response!

Erik Simonetti

I think enabling the sysmem fallback option solved it, it's training now. Thanks! ps: please keep making videos, your tutorials are some of the best and certainly the most entertaining ones by far!

Timmy

So I figured out how to keep the command window from closing right away. This is the error I am getting from the one click downloader: > ai-toolkit-ui@0.1.0 build_and_start > npm install && npm run update_db && npm run build && npm run start npm WARN EBADENGINE Unsupported engine { npm WARN EBADENGINE package: '@prisma/client@6.3.1', npm WARN EBADENGINE required: { node: '>=18.18' }, npm WARN EBADENGINE current: { node: 'v17.7.1', npm: '8.5.2' } npm WARN EBADENGINE } npm WARN EBADENGINE Unsupported engine { npm WARN EBADENGINE package: 'concurrently@9.1.2', npm WARN EBADENGINE required: { node: '>=18' }, npm WARN EBADENGINE current: { node: 'v17.7.1', npm: '8.5.2' } npm WARN EBADENGINE } npm WARN EBADENGINE Unsupported engine { npm WARN EBADENGINE package: 'next@15.1.7', npm WARN EBADENGINE required: { node: '^18.18.0 || ^19.8.0 || >= 20.0.0' }, npm WARN EBADENGINE current: { node: 'v17.7.1', npm: '8.5.2' } npm WARN EBADENGINE } npm WARN EBADENGINE Unsupported engine { npm WARN EBADENGINE package: 'prisma@6.3.1', npm WARN EBADENGINE required: { node: '>=18.18' }, npm WARN EBADENGINE current: { node: 'v17.7.1', npm: '8.5.2' } npm WARN EBADENGINE } npm WARN deprecated inflight@1.0.6: This module is not supported, and leaks memory. Do not use it. Check out lru-cache if you want a good and tested way to coalesce async requests by a key value, which is much more comprehensive and powerful. npm WARN deprecated @npmcli/move-file@1.1.2: This functionality has been moved to @npmcli/fs npm WARN deprecated npmlog@6.0.2: This package is no longer supported. npm WARN deprecated rimraf@2.7.1: Rimraf versions prior to v4 are no longer supported npm WARN deprecated rimraf@3.0.2: Rimraf versions prior to v4 are no longer supported npm WARN deprecated glob@7.2.3: Glob versions prior to v9 are no longer supported npm WARN deprecated glob@7.2.3: Glob versions prior to v9 are no longer supported npm WARN deprecated glob@7.2.3: Glob versions prior to v9 are no longer supported npm WARN deprecated glob@7.2.3: Glob versions prior to v9 are no longer supported npm WARN deprecated are-we-there-yet@3.0.1: This package is no longer supported. npm WARN deprecated gauge@4.0.4: This package is no longer supported. npm ERR! code 1 npm ERR! path H:\AI_TRAINING\Wan_2.1\Aitrepreneur\ai-toolkit\ui\node_modules\prisma npm ERR! command failed npm ERR! command C:\WINDOWS\system32\cmd.exe /d /s /c node scripts/preinstall-entry.js npm ERR! ┌──────────────────────────────────────────────┐ npm ERR! │ Prisma only supports Node.js >= 18.18. │ npm ERR! │ Please upgrade your Node.js version. │ npm ERR! └──────────────────────────────────────────────┘

Aitrepreneur

don't abandon it, let's try to find a solution. You can dm me

Vato Vegas

Thanks Timmy going to give this a try!

Testtn

Yes you pay for network storage, about $10 for 150Gb. Unfortunately it fills up fast with temp files that you can't even see. I'm not an expert so every week or so I would have to create a new at work storage and then copy my Loras and whatever I did not want to lose using runpodctl send and receive. If you search there docs under transferring files it's pretty straightforward and easy to use. I suggest when setting up a network storage you choose the Euro RO1 Romania server. It has the most GPUs available. Not every storage facility has the same access to GPUs. A little thing pops up on the side and you can see the different GPUs available. And don't forget to pay because they will just delete your storage if you don't .

Timmy

Also, as a side note, if you open a CMD window THEN run the installer batch directly from the command line, it won't close immediately if there is an error, and you will be able to read what caused the failure. That is how I figured out that it was Node.js on my end.

JAV Pomoe

Hello,I tried to run "AI-TOOLKIT_AUTO_INSTALL-RUNPOD.sh" using RunPod as instructed, but it failed as follows: === AI-Toolkit first-time installation === Which GPU is this pod running? [1] RTX-5000-series (Blackwell, sm_120) or newer [2] Anything older (Ada, Hopper, Ampere, etc.) Enter 1 or 2: 1 Get:1 http://archive.ubuntu.com/ubuntu jammy InRelease [270 kB] Get:2 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64 InRelease [1581 B] Get:3 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64 Packages [1851 kB] Err:3 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64 Packages File has unexpected size (1852870 != 1850507). Mirror sync in progress? [IP: 92.123.102.105 443] Hashes of expected file: - Filesize:1850507 [weak] - SHA256:97960e9159be087be063fd3fcd63217a13dcd40b7ae329f38ddcee2541e2406b - SHA1:7baa6fed520300c1fdb5eb10ca02367f0806deb9 [weak] - MD5Sum:4f4f50310c11c9a0ba217aacc75d8f47 [weak] Release file created at: Tue, 22 Jul 2025 19:05:10 +0000 Get:4 http://security.ubuntu.com/ubuntu jammy-security InRelease [129 kB] Get:5 https://ppa.launchpadcontent.net/deadsnakes/ppa/ubuntu jammy InRelease [18.1 kB] Get:6 http://archive.ubuntu.com/ubuntu jammy-updates InRelease [128 kB] Get:7 http://archive.ubuntu.com/ubuntu jammy-backports InRelease [127 kB] Get:8 http://archive.ubuntu.com/ubuntu jammy/universe amd64 Packages [17.5 MB] Get:9 https://ppa.launchpadcontent.net/deadsnakes/ppa/ubuntu jammy/main amd64 Packages [32.9 kB] Get:10 http://archive.ubuntu.com/ubuntu jammy/multiverse amd64 Packages [266 kB] Get:11 http://archive.ubuntu.com/ubuntu jammy/restricted amd64 Packages [164 kB] Get:12 http://archive.ubuntu.com/ubuntu jammy/main amd64 Packages [1792 kB] Get:13 http://security.ubuntu.com/ubuntu jammy-security/multiverse amd64 Packages [48.5 kB] Get:14 http://archive.ubuntu.com/ubuntu jammy-updates/restricted amd64 Packages [5163 kB] Get:15 http://archive.ubuntu.com/ubuntu jammy-updates/universe amd64 Packages [1574 kB] Get:16 http://archive.ubuntu.com/ubuntu jammy-updates/main amd64 Packages [3471 kB] Get:17 http://security.ubuntu.com/ubuntu jammy-security/main amd64 Packages [3160 kB] Get:18 http://archive.ubuntu.com/ubuntu jammy-updates/multiverse amd64 Packages [75.9 kB] Get:19 http://archive.ubuntu.com/ubuntu jammy-backports/main amd64 Packages [83.2 kB] Get:20 http://archive.ubuntu.com/ubuntu jammy-backports/universe amd64 Packages [35.2 kB] Get:21 http://security.ubuntu.com/ubuntu jammy-security/universe amd64 Packages [1269 kB] Get:22 http://security.ubuntu.com/ubuntu jammy-security/restricted amd64 Packages [4976 kB] Fetched 40.3 MB in 2s (22.6 MB/s) Reading package lists... Done E: Failed to fetch https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/Packages.gz File has unexpected size (1852870 != 1850507). Mirror sync in progress? [IP: 92.123.102.105 443] Hashes of expected file: - Filesize:1850507 [weak] - SHA256:97960e9159be087be063fd3fcd63217a13dcd40b7ae329f38ddcee2541e2406b - SHA1:7baa6fed520300c1fdb5eb10ca02367f0806deb9 [weak] - MD5Sum:4f4f50310c11c9a0ba217aacc75d8f47 [weak] Release file created at: Tue, 22 Jul 2025 19:05:10 +0000 E: Some index files failed to download. They have been ignored, or old ones used instead.

Timmy

Does anyone know how to use this with tensorflow / tensorboard?

Timmy

It sounds like the location of your Python folder was not added to the PATH variable in windows. Also, I have heard reports that the Microsoft Store version of Python can sometimes cause issues. I would download Python 3.12+ from this site directly and install it: https://www.python.org/downloads/ And when going through he setup process, make sure "Add to PATH" option is selected during setup. As a side note, you can also manually add the python path to your PATH variable in Windows. Just google "Add python path to PATH variable in windows" and it can step you through it better than I can here.

Bob Winberry

So, I downloaded the above, click the "Desktop" created a new folder in a different drive, "AITOOLS", ran the "AI-TOOLKIT_AUTO_INSTALL.bat" it loaded everythig, I put in my TXT and PNG into data set, watched the video again, set the settings, yet still get "Traceback (most recent call last): File "E:\AITOOLS\ai-toolkit\run.py", line 6, in from dotenv import load_dotenv ModuleNotFoundError: No module named 'dotenv' (venv) E:\AITOOLS\ai-toolkit>

J.ruthless

Appreciate your help Timmy this worked for me. And thanks for providing the link. Your a gentleman and a scholar. :)

Aitrepreneur

there might have been a connection issue because it's working for me now, maybe try again now

Aitrepreneur

Correct altough I recommend python 3.10.11 instead since most ai tools work with python 3.10 version

Aitrepreneur

dotenv error is indeed an issue caused by the absence of the Desktop development with C++ Make also sure that you have python 3.11.10 and that it is added to path and that you have the latest version of nodejs installed + the correct cuda toolkit for your GPU. Then you can also try going inside the folder install, click on the folder path type cmd press enter, this will open a cmd window inside the folder and inside type: venv\Scripts\activate pip install -r requirements Press enter and then wait until everything is installed then relaunch the launcher

Vato Vegas

Timmy this was totally it Thanks x 1,000! A Gentle Scholar like J Ruthless said! Thanks Bro!

LW

.... from an earlier thread.... K wrote: dotenv error is indeed an issue caused by the absence of the Desktop development with C++ Make also sure that you have python 3.11.10 and that it is added to path and that you have the latest version of nodejs installed + the correct cuda toolkit for your GPU. Then you can also try going inside the folder install, click on the folder path type cmd press enter, this will open a cmd window inside the folder and inside type: venv\Scripts\activate pip install -r requirements Press enter and then wait until everything is installed then relaunch the launcher Now you had me going for a while there K! I eventually worked out that it should be: pip install -r requirements.txt So, I ran that and waited for the update and then relaunched the launcher (LAUNCHER-TOOLKIT.bat) .... and................ Traceback (most recent call last): File "X:\WANimageloratraining3\ai-toolkit\run.py", line 6, in from dotenv import load_dotenv ModuleNotFoundError: No module named 'dotenv' (venv) X:\WANimageloratraining3\ai-toolkit> I have CUDA 12.6 installed and in PATH (I'm using a 3090) I have Python 3.13.5 installed and in PATH I have Desktop development with C++ installed I have the latest nodejs installed I have tried a complete new install with the one-click installer (AI-TOOLKIT_AUTO_INSTALL.bat) but always the same dotenv error when trying to run a job. Any suggestions?

LW

btw... I do see the following message.... File "C:\Python313\Lib\subprocess.py", line 1554, in _execute_child hp, ht, pid, tid = _winapi.CreateProcess(executable, args, ~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^ # no special security ^^^^^^^^^^^^^^^^^^^^^ ...<4 lines>... cwd, ^^^^ startupinfo) ^^^^^^^^^^^^ FileNotFoundError: [WinError 2] The system cannot find the file specified [end of output] note: This error originates from a subprocess, and is likely not a problem with pip. error: subprocess-exited-with-error × Getting requirements to build wheel did not run successfully. │ exit code: 1 ╰─> See above for output. note: This error originates from a subprocess, and is likely not a problem with pip. and when launching AI-toolkit: Starting AIÔÇæToolkit UI... (Press Ctrl+C to stop) > ai-toolkit-ui@0.1.0 build_and_start > npm install && npm run update_db && npm run build && npm run start up to date, audited 446 packages in 1s 61 packages are looking for funding run `npm fund` for details 5 vulnerabilities (1 low, 2 high, 2 critical) To address issues that do not require attention, run: npm audit fix To address all issues, run: npm audit fix --force I did try to run npm audit fix --force but that didn't help

Aitrepreneur

yeah you need the python 3.10, preferably the 3.10.11 added to path

LW

ok, so I installed python 3.10.11 and added it to PATH But still got the same error even after restarting the computer. Traceback (most recent call last): File "X:\WANimageloratraining3\ai-toolkit\run.py", line 6, in from dotenv import load_dotenv ModuleNotFoundError: No module named 'dotenv' I still see the following on running the AI-toolkit laucher:- > ai-toolkit-ui@0.1.0 build_and_start > npm install && npm run update_db && npm run build && npm run start up to date, audited 446 packages in 4s 61 packages are looking for funding run `npm fund` for details 5 vulnerabilities (1 low, 2 high, 2 critical) Is this part of the problem, or unrelated?

Wilhelm Zabłotnikov

Are you planning to release some nice workflow for Wan2.2 or something?

Rule801

Currently using Automatic1111 SD , run locally . Will i need to have Comfy UI installed before I run the most recent Ai Toolkit/Wan installer ?

Timmy

See the video posted above for more details: Tensorboard venv Install command (Inside the venv): python.exe -m pip install tensorboard Triton for windows install command (Inside the venv): python.exe -m pip install triton-windows jobConfig.ts Location in AI Toolkit folder: "ai_toolkit\ui\src\app\jobs\new\jobConfig.ts" Tensorboard lines to add to jobConfig.ts: log_dir: '.\\run', logging: { log_every: 2, use_wandb: false, verbose: false, }, Tensorboard Launcher: @echo off :: Activate Python virtual environment if not exist "venv\Scripts\activate.bat" ( echo [ERROR] venv not found. Run the installer first. pause exit /b 1 ) call "venv\Scripts\activate.bat" :: Launch TensorBoard with the 'run' directory as the log directory tensorboard --logdir=.\run --host=localhost --port=6006

Rui Holdorf

can you post the dataset you used to train Margot LoRA, please?

Vyacheslav

Hi! Wonderful workflow! When I'm trying to do img2video, for some reason, video doesn't take my first frame, but creating a new one, accordingly to the prompt I put.

Aitrepreneur

Make sure that you are using the image to video model

Hans Velasquez

I tried installing the AI Toolkit but whenever I click Launcher-Toolkit.bat, it just opens Terminal for a split second and then disappears...

Aitrepreneur

open a cmd window and drag and drop the Launcher-Toolkit.bat inside then press enter, it will at least give us an error we can use to see what's wrong

DavidO

I have similar issue, does this help - Microsoft Windows [Version 10.0.26100.2605] (c) Microsoft Corporation. All rights reserved. D:\AITOOLKIT\ai-toolkit>D:\AITOOLKIT\ai-toolkit\LAUNCHER-TOOLKIT.bat Starting AIÔÇæToolkit UI... (Press Ctrl+C to stop) 'npm' is not recognized as an internal or external command, operable program or batch file. D:\AITOOLKIT\ai-toolkit>

DavidO

Edit I havn't installed - Node.js 18+ so disregard last please, I will circle round to this again when time allows

Boxertwin

Mine does the same I get this when I drop the launcher into cmd: "[ERROR] venv not found. Run the installer first." Am I forgetting to do something obvious?

mikki

Are we going to be able to have wan 2.2 config on Aitoolkit locally here soon?

#DFined

So, while the training seems to go great, the samples become good quite quickly, when I try to use the lora in the Wan t2i workflows, the result looks nothing like the samples - the effects are greatly diminished and often barely visible. For character loras it works fine, but for more complex concept loras - I keep running into this problem. Like a tried a battle-scarred-look lora and it worked great in the training samples, but in the workflows, the effects were much much weaker. I tried fiddling with samplers, cfg, etc, but still no luck. And the training itself seems to be working, as the samples do look good, but it just doesn't work in-workflow later. Do I need to train on a different model than just the base one or something?

Aitrepreneur

no but check that you are running the correct model, keep in mind that if you use other loras than lightx2v it might change the final result. You can also increase the weight of the lora as well

Patrick Brown

I did this last night and trained a Lora for Lake Bell and it turned out amazing! Thank you! However, today, I'm receiving an error when I try to train a new Lora. The error message is: "An error occurred while downloading using `hf_transfer`. Consider disabling HF_HUB_ENABLE_HF_TRANSFER for better error handling." I've tried multiple ways to disable this including setting this variable to '0' in the terminal as well as with a python script and can't seem to get it to download the files without erroring out. I even nuked my pod and created a brand new one and am getting the same error. Any idea for how to disable this so I can continue my training?

Aitrepreneur

this error actually happens because you don't have enough space on the pod

Patrick Brown

Crazy. I just stood it up. Thank you!! I’ll go tweak it and try again.

Wicked Desires

I'm getting this error, "venv not found. Run the installer first". venv is clearly in the folder. I checked to make sure venv\Scripts\activate.bat is avaialbe. It is also available.

Ghost

I downloaded your old install for flux kontext for AI toolkit, can I somehow update that installation without breaking it and if so how? Seems like a huge waste of bandwidth to redownload the entire thing again and have to delete the old one to use this

Gerard Pregenzer

Dumb question from someone who is completely new to all of this. Is there a way to use the AI toolkit to train a LoRA through Pixel Dojo AI?

Mango

Beginner question, How do i get FusionX as a .gguf. not missing any files and everything updated.

Jack

For Wan 2.2 I've manage to train t2v, for Wan 2.2 I2v what are your recomendations for the regularization images?

SAMIRA

I want to train NSFW but for the descriptions, GPT chat doesn't give them to me.

D Dunn

getting this error on runpod CUDA out of memory. Tried to allocate 68.00 MiB. GPU 0 has a total capacity of 31.37 GiB of which 58.69 MiB is free. Including non-PyTorch memory, this process has 31.30 GiB memory in use. Of the allocated memory 30.57 GiB is allocated by PyTorch, and 144.35 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)

James Inage

Do you know if we can do 2.2 yet? Thanks for your time

Kent Umeki

will i have better results if i train the model using videos? or does it not matter?

Dauron Robinson

I've had luck with using gemini for nsfw captions/descriptions of images

totally_not_a_weirdo

"[ERROR] curl not found ΓÇô install Windows 10 v1803+ or add it to PATH." This pops up immediately after choosing which RTX-series card (on the installer). I definitely have curl.exe on my computer, and it sure seems to be added to PATH. ETA> After over 2 years of obsessing over this stuff, today I learned that I have been doing the PATH thing wrong. Having corrected that (those, actually), it installed as expected.

Matthew Smallwood

I'm trying to use this on runpod with an RTX pro 6000. i get this error. "Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!"

nurtu

CUDA error: misaligned address CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1 Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.

newtoai

Hi. Could you please fix the auto installation for offline use ?