Skip to content

Issues: ostris/ai-toolkit

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

75 Open 113 Closed
75 Open 113 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Doesn't work. Errors out when I use the Python web UI and click "Start Training" , with "cannot import 'packaging' form 'pkg_resources'
#228 opened Nov 25, 2024 by nikolatesla20
5
Slider training not working
#227 opened Nov 24, 2024 by Volnovik
Iteration speed slows down exponentially
#225 opened Nov 20, 2024 by orionflame
how to load the flux model after finetune using ai-toolkit? In other words, how to merge lora checkpoint to the original flux dev model?
#223 opened Nov 19, 2024 by MathamPollard
How can we use a model already downloaded? Instead of using huggingface?
#222 opened Nov 18, 2024 by moreAImore
Error running job: expected a valid token but found ':' here:
#221 opened Nov 16, 2024 by StartHua
The parameter type of batch in TrainSliderProcess.py does not match the receiving type.
#219 opened Nov 12, 2024 by Ando-Lin
1
how to save the loss during lora finetune
#218 opened Nov 12, 2024 by MathamPollard
1
How to Increase the Maximum Input Length for flux-schnell Model？
#217 opened Nov 11, 2024 by mocorr
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!
#215 opened Nov 8, 2024 by SAC020
1
How to merge trained FLUX LoRA into FLUX Dev Model
#213 opened Nov 7, 2024 by wiggin66
3
Flux trainer autocaption inconsistency
#211 opened Nov 5, 2024 by achuinard
1
can I specify that the image resolution ratio is 720x1440?
#210 opened Nov 5, 2024 by github-water
1
OSError: CUDA_HOME environment variable is not set. Please set it to your CUDA install root
#208 opened Nov 3, 2024 by immodalmul
Getting Error Again and Again
#206 opened Nov 1, 2024 by bokachodalawdsa
when i finished training the cuda mem is stilled occupied , how to free the mem?
#205 opened Oct 30, 2024 by alpttex19
1
Flux lora train selects Flux.1-dev, but it will download files such as text_encoder in FLUX.1-schnell instead of the corresponding files in Flux.1-dev
#204 opened Oct 26, 2024 by wen313
1
[cudnn_frontend] Error: No execution plans support the graph (--Solved--)
#203 opened Oct 25, 2024 by AfterHAL
SD 3.5 Loras still prints the wrong hash "e3b0c442..."
#202 opened Oct 24, 2024 by diodiogod
Using certain naming conventions for the directory path breaks loading previous saves for LoRA Training
#200 opened Oct 22, 2024 by cswattts
Hello, i have a training issue when loading on replicate for FLUX.
#198 opened Oct 15, 2024 by LIQUIDMIND111
AttributeError: module 'timm.models._registry' has no attribute 'get_pretrained_cfgs_for_arch'. Did you mean: 'get_pretrained_cfg_value'?
#197 opened Oct 15, 2024 by TrickyBarrel
2
Errors after new version release
#195 opened Oct 11, 2024 by pianogospel
15
Error running job: 'QLinear' object has no attribute 'qweight'
#193 opened Oct 9, 2024 by Joir-Jio
3
Misinterpretation of attention masking
#192 opened Oct 6, 2024 by AmericanPresidentJimmyCarter
1
ProTip! Add no:assignee to see everything that’s not assigned.