-
It might be an upstream issue.
I'm using Forge with default settings, except for the resolution. However, I’ve tried most of the samplers and schedulers to fix the problem, but without success. Wha…
-
### Please check that this issue hasn't been reported before.
- [X] I searched previous [Bug Reports](https://github.com/OpenAccess-AI-Collective/axolotl/labels/bug) didn't find any similar reports.
…
-
I tested some of the datasets in the eval set and found that the results are higher than those in the paper. Is there an update to the LoRA model compared to what was presented in the paper?
-
Hello.
I definitely can't get rid of this error.
it comes from the file :
File "ComfyUI_windows_portable\python_embeded\lib\site-packages\diffusers\loaders\lora.py", line 107, in load_lora_weigh…
-
### Checklist
- [X] The issue exists after disabling all extensions
- [X] The issue exists on a clean installation of webui
- [ ] The issue is caused by an extension, but I believe it is caused by a …
-
### System Info
NA
### Who can help?
@BenjaminBossan could you add a note to the docs to explain the default behaviour, and also any work arounds (e.g. loading a base model and dequantizing and loa…
-
### Anything you want to discuss about vllm.
I've fine-tuned Qwen2.5-14B-Instruct using QLora(bitsandbytes 4bit) and also a full fine-tune. However when I tried to use it with a quantized model (Qw…
-
### Your current environment
```text
podman --version
podman version 5.2.3
uname -a
Linux noelo-work 6.10.12-200.fc40.x86_64 #1 SMP PREEMPT_DYNAMIC Mon Sep 30 21:38:25 UTC 2024 x86_64 GNU/L…
-
!swift sft \
--model_type got-ocr2 \
--model_id_or_path stepfun-ai/GOT-OCR2_0 \
--sft_type lora \
--dataset /kaggle/working/output_data.json \
--output_dir /kaggle/working/hindi_got_model_3 \
--…
-
### Issue Description
Trying to use this lora on Flux and it's not loading. Lora was trained on civitai with their default (as of now) training setup
I don't know if the Flux "flavor" (i.e. qint…