repo stringclasses 1
value | github_id int64 1.27B 4.95B | github_node_id stringlengths 18 24 | number int64 8 14.3k | html_url stringlengths 49 53 | api_url stringlengths 59 63 | title stringlengths 1 402 | body stringlengths 1 62.9k ⌀ | state stringclasses 2
values | state_reason stringclasses 4
values | locked bool 2
classes | comments_count int64 0 99 | labels listlengths 0 5 | assignees listlengths 0 5 | created_at stringdate 2022-06-09 16:28:35 2026-07-22 10:32:58 | updated_at stringdate 2022-06-12 22:18:01 2026-07-22 10:32:58 | closed_at stringdate 2022-06-12 22:18:01 2026-07-21 13:17:28 ⌀ | author_association stringclasses 3
values | milestone_title stringclasses 0
values | snapshot_id stringclasses 149
values | extracted_at stringdate 2026-04-07 13:34:13 2026-07-22 10:35:16 | author_login stringlengths 3 28 | author_id int64 1.54k 296M | author_node_id stringlengths 12 20 | author_type stringclasses 3
values | author_site_admin bool 1
class |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
huggingface/diffusers | 1,473,511,594 | I_kwDOHa8MBc5X1ACq | 1,530 | https://github.com/huggingface/diffusers/issues/1530 | https://api.github.com/repos/huggingface/diffusers/issues/1530 | `VQDiffusionPipeline` missing `learned_classifier_free_sampling_embeddings` | ### Describe the bug
When trying to load a pretrained `VQDiffusionPipeline`, it says that `learned_classifier_free_sampling_embeddings` is not passed. I also looked at the cache directory, and the corresponding folder is also missing (https://huggingface.co/microsoft/vq-diffusion-ithq/tree/main/learned_classifier_free... | closed | completed | false | 5 | [
"bug",
"stale"
] | [
"williamberman"
] | 2022-12-02T22:04:08Z | 2023-01-02T15:05:39Z | 2023-01-02T15:05:39Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | xenova | 26,504,141 | MDQ6VXNlcjI2NTA0MTQx | User | false |
huggingface/diffusers | 1,474,267,653 | I_kwDOHa8MBc5X34oF | 1,534 | https://github.com/huggingface/diffusers/issues/1534 | https://api.github.com/repos/huggingface/diffusers/issues/1534 | Error in Dreambooth Training notebook | ### Describe the bug
Error in step 50/300 on 'run training' cell in [Dreambooth Training Colab](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/sd_dreambooth_training.ipynb) notebook
### Reproduction
_No response_
### Logs
```shell
----------------------------------------------... | closed | completed | false | 2 | [
"bug"
] | [] | 2022-12-03T23:00:52Z | 2022-12-05T18:32:03Z | 2022-12-05T18:32:03Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | estelleflores | 83,784,489 | MDQ6VXNlcjgzNzg0NDg5 | User | false |
huggingface/diffusers | 1,474,296,456 | I_kwDOHa8MBc5X3_qI | 1,535 | https://github.com/huggingface/diffusers/issues/1535 | https://api.github.com/repos/huggingface/diffusers/issues/1535 | Image Variation Pipeline | ### Describe the bug
Image variation pipeline is giving me the following error:
` File "/home/ubuntu/.local/lib/python3.8/site-packages/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_image_variation.py", line 401, in __call__
image_embeddings = self._encode_image(image, device, num_images_per_p... | closed | completed | false | 5 | [
"bug",
"stale"
] | [] | 2022-12-04T00:36:00Z | 2023-01-13T15:04:22Z | 2023-01-13T15:04:22Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | misbahsy | 35,427,888 | MDQ6VXNlcjM1NDI3ODg4 | User | false |
huggingface/diffusers | 1,474,386,720 | I_kwDOHa8MBc5X4Vsg | 1,536 | https://github.com/huggingface/diffusers/issues/1536 | https://api.github.com/repos/huggingface/diffusers/issues/1536 | Custom pipeline : lpw_stable_diffusion is incompatible with new version of diffusers | ### Describe the bug
```
DiffusionPipeline.from_pretrained(
model_id, custom_pipeline="lpw_stable_diffusion",
torch_dtype=torch.float16, revision="fp16",
safety_checker=None, vae=vae, use_auth_token=hf_token
).to("cuda")
```
Gives this error
mod... | closed | completed | false | 3 | [
"bug"
] | [] | 2022-12-04T03:11:49Z | 2022-12-09T03:52:58Z | 2022-12-07T10:20:23Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | adhikjoshi | 11,740,719 | MDQ6VXNlcjExNzQwNzE5 | User | false |
huggingface/diffusers | 1,474,453,168 | I_kwDOHa8MBc5X4l6w | 1,537 | https://github.com/huggingface/diffusers/issues/1537 | https://api.github.com/repos/huggingface/diffusers/issues/1537 | Apple MPS and KDPM2AncestralDiscreteScheduler & KDPM2DiscreteScheduler | ### Describe the bug
The above two schedulers have issues when running under Apple MPS. It looks as if `float64` to `float32` conversion is not handled correctly for MPS for either of these. See attached screenshots for errors.
![CleanShot 2022-12-04 at 08 32 41@2x](https://user-images.githubusercontent.com/181110/2... | closed | completed | false | 4 | [
"bug",
"stale"
] | [
"pcuenca"
] | 2022-12-04T04:53:44Z | 2023-02-05T22:45:17Z | 2023-02-05T22:45:17Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | FahimF | 181,110 | MDQ6VXNlcjE4MTExMA== | User | false |
huggingface/diffusers | 1,474,698,260 | I_kwDOHa8MBc5X5hwU | 1,538 | https://github.com/huggingface/diffusers/issues/1538 | https://api.github.com/repos/huggingface/diffusers/issues/1538 | StableDiffusionMegaPipeline Not Working with stabilityai/stable-diffusion-2-base | ### Describe the bug
custom_pipeline="stable_diffusion_mega" used with stable diffusion 2 not working
```
pipe = DiffusionPipeline.from_pretrained(
"stabilityai/stable-diffusion-2-base",
custom_pipeline="stable_diffusion_mega",
torch_dtype=torch.float16, revision... | closed | completed | false | 5 | [
"bug",
"stale"
] | [] | 2022-12-04T10:46:50Z | 2023-01-11T15:05:07Z | 2023-01-11T15:05:07Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | adhikjoshi | 11,740,719 | MDQ6VXNlcjExNzQwNzE5 | User | false |
huggingface/diffusers | 1,474,710,933 | I_kwDOHa8MBc5X5k2V | 1,539 | https://github.com/huggingface/diffusers/issues/1539 | https://api.github.com/repos/huggingface/diffusers/issues/1539 | Conversion Script Type Int Incorrect | ### Describe the bug
In https://github.com/huggingface/diffusers/blob/main/scripts/convert_original_stable_diffusion_to_diffusers.py
The argument --prediction_type is incorrectly defined as INT when the helper is guiding to pass type STRING.
### Reproduction
Any attempt to use the script to convert diffusers whe... | closed | completed | false | 1 | [
"bug"
] | [] | 2022-12-04T11:04:01Z | 2022-12-05T18:51:56Z | 2022-12-05T18:51:56Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | DrewWalkup | 44,742,697 | MDQ6VXNlcjQ0NzQyNjk3 | User | false |
huggingface/diffusers | 1,474,781,117 | I_kwDOHa8MBc5X51-9 | 1,540 | https://github.com/huggingface/diffusers/issues/1540 | https://api.github.com/repos/huggingface/diffusers/issues/1540 | Dreambooth / transformers cannot execute, missing "MODEL_TYPE" | ### Describe the bug
When attempting to execute dreambooth on any version of transformers >4.21 run into the issue outlined in LOGS with vanilla SDv1.5.
### Reproduction
accelerate launch --mixed_precision='fp16' train_dreambooth.py \
--train_text_encoder \
--save_steps=500 \
--pretrained_model_na... | closed | completed | false | 3 | [
"bug",
"stale"
] | [] | 2022-12-04T12:39:12Z | 2023-01-11T15:05:06Z | 2023-01-11T15:05:06Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | DrewWalkup | 44,742,697 | MDQ6VXNlcjQ0NzQyNjk3 | User | false |
huggingface/diffusers | 1,474,800,888 | I_kwDOHa8MBc5X56z4 | 1,541 | https://github.com/huggingface/diffusers/issues/1541 | https://api.github.com/repos/huggingface/diffusers/issues/1541 | Dreambooth cannot determine MODEL_TYPE, CKPTs do not load | ### Describe the bug
Dreambooth cannot determine what MODEL_TYPE Stable Diffusion v1.5 is and, if one manages to producing a training from the script in spit of the issues, the resulting diffusers are unusable due to TYPE being NONE.
### Reproduction
Complete a dreambooth training:
accelerate launch --mixed_precisi... | closed | completed | false | 14 | [
"bug",
"stale"
] | [] | 2022-12-04T13:11:28Z | 2023-01-14T15:03:29Z | 2023-01-14T15:03:29Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | DrewWalkup | 44,742,697 | MDQ6VXNlcjQ0NzQyNjk3 | User | false |
huggingface/diffusers | 1,474,819,215 | I_kwDOHa8MBc5X5_SP | 1,542 | https://github.com/huggingface/diffusers/issues/1542 | https://api.github.com/repos/huggingface/diffusers/issues/1542 | Error with StableDiffusionPipeline from diffusers | here is the version i'm using
the code cell with the problem
and here is the error message... | closed | completed | false | 2 | [] | [] | 2022-12-04T13:38:33Z | 2022-12-04T15:11:13Z | 2022-12-04T15:11:13Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | alielfilali01 | 58,257,628 | MDQ6VXNlcjU4MjU3NjI4 | User | false |
huggingface/diffusers | 1,474,848,087 | I_kwDOHa8MBc5X6GVX | 1,543 | https://github.com/huggingface/diffusers/issues/1543 | https://api.github.com/repos/huggingface/diffusers/issues/1543 | textual_inversion.py - Cannot set revision="fp16" | ### Describe the bug
Errors with RuntimeError: Input type (float) and bias type (c10::Half) should be the same
if attempt to set revision="fp16"
### Reproduction
examples/textual_inversion.py errors if revision set to fp16
### Logs
```shell
Traceback (most recent call last):
File "textual_inversion.py", line 6... | closed | completed | false | 5 | [
"bug",
"stale"
] | [
"williamberman",
"patil-suraj"
] | 2022-12-04T14:04:24Z | 2023-01-03T19:37:10Z | 2023-01-03T19:37:10Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | zagglez | 61,600,248 | MDQ6VXNlcjYxNjAwMjQ4 | User | false |
huggingface/diffusers | 1,475,247,236 | I_kwDOHa8MBc5X7nyE | 1,544 | https://github.com/huggingface/diffusers/issues/1544 | https://api.github.com/repos/huggingface/diffusers/issues/1544 | AttributeError: 'DDPMScheduler' object has no attribute 'get_velocity' (using stable-diffusion-2) | ### Describe the bug
Running `examples/dreambooth/train_dreambooth.py` fails:
```
[noob@server dreambooth]$ ./train.sh
You are using a model of type clip_text_model to instantiate a model of type . This is not supported for all configurations of models and can yield errors.
Steps: 0%| ... | closed | completed | false | 7 | [
"bug",
"stale"
] | [] | 2022-12-04T22:23:26Z | 2023-03-07T22:51:36Z | 2023-01-12T15:04:29Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | Dr-Noob | 22,979,883 | MDQ6VXNlcjIyOTc5ODgz | User | false |
huggingface/diffusers | 1,475,966,073 | I_kwDOHa8MBc5X-XR5 | 1,545 | https://github.com/huggingface/diffusers/issues/1545 | https://api.github.com/repos/huggingface/diffusers/issues/1545 | Generating artifacts when changing the schedulers to DDIM and DDPM | Hello. I have tried to change and test different schedulers after loading the model as described in the following page: https://huggingface.co/docs/diffusers/using-diffusers/schedulers
Though the DMP and PNDM give acceptable results, there are some issues with the DDIM and DDPM schedulers as shown in the results below... | closed | completed | false | 3 | [] | [] | 2022-12-05T08:09:05Z | 2022-12-15T01:39:07Z | 2022-12-15T01:39:07Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | abassino | 10,333,061 | MDQ6VXNlcjEwMzMzMDYx | User | false |
huggingface/diffusers | 1,476,095,080 | I_kwDOHa8MBc5X-2xo | 1,546 | https://github.com/huggingface/diffusers/issues/1546 | https://api.github.com/repos/huggingface/diffusers/issues/1546 | Truncation warning for StableDiffusionPipeline causes error for batched inputs | ### Describe the bug
When running StableDiffusionPipeline with a batch of inputs of different lengths it seems there will now be an error.
```
ValueError: Unable to create tensor, you should probably activate truncation and/or padding with 'padding=True' 'truncation=True' to have batched tensors with the same leng... | closed | completed | false | 6 | [
"bug",
"stale"
] | [] | 2022-12-05T09:20:41Z | 2023-01-14T15:03:27Z | 2023-01-14T15:03:27Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | coleridge72 | 35,977,965 | MDQ6VXNlcjM1OTc3OTY1 | User | false |
huggingface/diffusers | 1,476,841,803 | I_kwDOHa8MBc5YBtFL | 1,551 | https://github.com/huggingface/diffusers/issues/1551 | https://api.github.com/repos/huggingface/diffusers/issues/1551 | Dimension of `encoder_hidden_state` in UNet2DConditionModel | I'm trying to train my own img2img pipeline guided by a custom text embedding. Currently, I'm using diffusion on pixel space directly. When I looked at the inputs to the [call function of the StableDiffusionImg2ImgPipeline](https://github.com/huggingface/diffusers/blob/6b02323a602a66841729c3a5d60844b24aa81ff2/src/diffu... | closed | completed | false | 4 | [] | [] | 2022-12-05T15:40:55Z | 2022-12-06T13:36:10Z | 2022-12-06T13:36:09Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | AlienKevin | 22,850,071 | MDQ6VXNlcjIyODUwMDcx | User | false |
huggingface/diffusers | 1,476,905,838 | I_kwDOHa8MBc5YB8tu | 1,552 | https://github.com/huggingface/diffusers/issues/1552 | https://api.github.com/repos/huggingface/diffusers/issues/1552 | Stable Diffusion 2 doesn't work with `enable_attention_slicing()` | ### Describe the bug
As pointed out here: https://github.com/huggingface/diffusers/issues/1484#issuecomment-1335562598 attention slicing doesn't work with SD 2.
### Reproduction
```python
import torch
from diffusers import StableDiffusionPipeline, EulerDiscreteScheduler
model_id = "stabilityai/stable-diffusion-... | closed | completed | false | 1 | [
"bug"
] | [] | 2022-12-05T16:13:54Z | 2022-12-08T18:50:01Z | 2022-12-08T18:50:01Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | patrickvonplaten | 23,423,619 | MDQ6VXNlcjIzNDIzNjE5 | User | false |
huggingface/diffusers | 1,477,062,200 | I_kwDOHa8MBc5YCi44 | 1,556 | https://github.com/huggingface/diffusers/issues/1556 | https://api.github.com/repos/huggingface/diffusers/issues/1556 | Images are black when only using GPU (CUDA 11.7) along with NSFW warning | ### Describe the bug
I am new to AI in general, and I am trying to use the diffusers package by following the hugging face tutorial for running the model online and locally. But both of the methods (GPU) gives me a black image in return. However when I use the CPU mode `pipe.to("cpu")`, I get the results as expected.
... | closed | completed | false | 12 | [
"bug"
] | [] | 2022-12-05T17:31:58Z | 2025-09-12T03:28:57Z | 2022-12-06T16:31:14Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | 1MochaChan1 | 74,943,095 | MDQ6VXNlcjc0OTQzMDk1 | User | false |
huggingface/diffusers | 1,478,111,945 | I_kwDOHa8MBc5YGjLJ | 1,565 | https://github.com/huggingface/diffusers/issues/1565 | https://api.github.com/repos/huggingface/diffusers/issues/1565 | When use `stabilityai/stable-diffusion-2`, it always use GPU:0, how can I use specify GPU - Out of memory | ### Describe the bug
I found that if I use `CompVis/stable-diffusion-v1-4`，it is ok.But when I use `stabilityai/stable-diffusion-2`,it has this problem. How can I deal it?
____________________________________________________________
when I train dreambooth, I use `stabilityai/stable-diffusion-2`. It always use GPU:... | closed | completed | false | 4 | [
"bug",
"stale"
] | [] | 2022-12-06T03:57:30Z | 2023-01-13T15:04:19Z | 2023-01-13T15:04:19Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | ZJJHYM | 55,906,730 | MDQ6VXNlcjU1OTA2NzMw | User | false |
huggingface/diffusers | 1,478,393,802 | I_kwDOHa8MBc5YHn_K | 1,566 | https://github.com/huggingface/diffusers/issues/1566 | https://api.github.com/repos/huggingface/diffusers/issues/1566 | Dreambooth: crash after saving a checkpoint if fp16 output is enabled | ### Describe the bug
If (`accelerate` is configured with `fp16`, or `--mixed_precision=fp16` is specified on the command line) AND `--save_steps` is specified on the command line, Dreambooth crashes after writing a checkpoint:
```bash
RuntimeError: Input type (torch.cuda.HalfTensor) and weight type (torch.cuda.Flo... | closed | completed | false | 1 | [
"bug"
] | [
"patil-suraj"
] | 2022-12-06T06:57:02Z | 2022-12-10T14:45:47Z | 2022-12-10T14:45:47Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | timh | 139,355 | MDQ6VXNlcjEzOTM1NQ== | User | false |
huggingface/diffusers | 1,479,244,091 | I_kwDOHa8MBc5YK3k7 | 1,569 | https://github.com/huggingface/diffusers/issues/1569 | https://api.github.com/repos/huggingface/diffusers/issues/1569 | Error in Dreambooth Inference Example Notebook | ### Describe the bug
Dreambooth inference notebook giving errors. This model was working on this notebook just a few days ago. Anyone has a suggestion on how to fix this?
### Reproduction
from torch import autocast
from diffusers import DiffusionPipeline
import torch
model_id = "sd-dreambooth-library/crisimse... | closed | completed | false | 3 | [
"bug"
] | [] | 2022-12-06T14:35:27Z | 2022-12-13T16:19:40Z | 2022-12-12T19:15:01Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | estelleflores | 83,784,489 | MDQ6VXNlcjgzNzg0NDg5 | User | false |
huggingface/diffusers | 1,479,678,229 | I_kwDOHa8MBc5YMhkV | 1,572 | https://github.com/huggingface/diffusers/issues/1572 | https://api.github.com/repos/huggingface/diffusers/issues/1572 | `stable_diffusion_mega` pipeline doesn't work for img2img and inpainting | ### Describe the bug
Hello, I runed the sample code of [stable-diffusion-mega](https://github.com/huggingface/diffusers/tree/main/examples/community#stable-diffusion-mega) and there's a mismatch between parameter names of new and legacy pipelines.
It seems that Mega pipeline uses `image` for `img2img` and `inpaint`... | closed | completed | false | 7 | [
"bug"
] | [] | 2022-12-06T17:18:43Z | 2022-12-07T06:25:18Z | 2022-12-07T06:25:18Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | juno-hwang | 52,326,857 | MDQ6VXNlcjUyMzI2ODU3 | User | false |
huggingface/diffusers | 1,479,803,912 | I_kwDOHa8MBc5YNAQI | 1,573 | https://github.com/huggingface/diffusers/issues/1573 | https://api.github.com/repos/huggingface/diffusers/issues/1573 | [Maintenance] Updating fast pipeline tests to rely on common `PipelineTesterMixin` | ### Context
The #1526 PR implemented a `PipelineTesterMixin` with common fast tests for pipelines to inherit: saving/loading, casting to fp16, checking tuple outputs, attention slicing, etc.
To support `PipelineTesterMixin`, a fast tests class has to:
1. Inherit form the mixin
https://github.com/huggingface/diffu... | closed | completed | false | 7 | [
"wip"
] | [
"patrickvonplaten",
"anton-l"
] | 2022-12-06T18:11:26Z | 2024-11-21T11:28:06Z | 2024-11-21T11:28:06Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | anton-l | 26,864,830 | MDQ6VXNlcjI2ODY0ODMw | User | false |
huggingface/diffusers | 1,479,848,583 | I_kwDOHa8MBc5YNLKH | 1,574 | https://github.com/huggingface/diffusers/issues/1574 | https://api.github.com/repos/huggingface/diffusers/issues/1574 | Dreambooth training continues to fail on bare metal Linux with 8GB GPU | ### Describe the bug
Both class image generation and training with pre-loaded images runs out of memory on Step 0.
### Reproduction
Here is the command being used:
```
export MODEL_NAME="stabilityai/stable-diffusion-2"
export INSTANCE_DIR="training/images/"
export CLASS_DIR="class/"
export OUTPUT_DIR="output... | closed | completed | false | 5 | [
"bug"
] | [] | 2022-12-06T18:29:47Z | 2022-12-07T03:44:17Z | 2022-12-06T20:25:06Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | VisionaryMind | 47,434,136 | MDQ6VXNlcjQ3NDM0MTM2 | User | false |
huggingface/diffusers | 1,480,351,601 | I_kwDOHa8MBc5YPF9x | 1,577 | https://github.com/huggingface/diffusers/issues/1577 | https://api.github.com/repos/huggingface/diffusers/issues/1577 | Inference is slower with SD2.0 and memory_efficient_attention | ### Describe the bug
The release notes claim that using `enable_xformers_memory_efficient_attention` makes inference faster.
In my tests, this is true when using SD 1.5, but false when using SD 2.0.
Some timings (in seconds) from my machine (using a Quadro RTX 8000), using Euler scheduler and 30 steps with the S... | closed | completed | false | 12 | [
"bug",
"stale"
] | [
"pcuenca"
] | 2022-12-06T22:31:40Z | 2024-07-16T11:37:41Z | 2023-01-17T15:04:02Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | antoche | 1,627,384 | MDQ6VXNlcjE2MjczODQ= | User | false |
huggingface/diffusers | 1,480,997,580 | I_kwDOHa8MBc5YRjrM | 1,578 | https://github.com/huggingface/diffusers/issues/1578 | https://api.github.com/repos/huggingface/diffusers/issues/1578 | Question about prior loss | Hello, in my training, the prior loss does not decrease and has been maintained at a certain value.
 is passed to `tokenizer.convert_tokens_to_ids()`. However, args.placeholder_token hasn't been tokeni... | closed | completed | false | 5 | [
"bug",
"stale"
] | [] | 2022-12-07T06:25:48Z | 2023-01-14T15:03:24Z | 2023-01-14T15:03:24Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | askerlee | 1,575,461 | MDQ6VXNlcjE1NzU0NjE= | User | false |
huggingface/diffusers | 1,481,121,370 | I_kwDOHa8MBc5YSB5a | 1,581 | https://github.com/huggingface/diffusers/issues/1581 | https://api.github.com/repos/huggingface/diffusers/issues/1581 | RePaint pipeline and stable-diffusion-inpaint pipeline legacy off-by-one error | ### Describe the bug
When I am using diffuser library to merge repaint pipeline logic with the SD-inpaint-pipeline-legacy, I observed an obvious performance drop which is very curious to me. I believe it is caused by two off-by-one errors.
The repaint paper introduced a pipeline to merge the known part of the image... | closed | completed | false | 6 | [
"bug",
"stale"
] | [
"anton-l"
] | 2022-12-07T06:37:38Z | 2023-02-08T15:03:21Z | 2023-02-08T15:03:20Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | Randolph-zeng | 11,933,185 | MDQ6VXNlcjExOTMzMTg1 | User | false |
huggingface/diffusers | 1,482,226,064 | I_kwDOHa8MBc5YWPmQ | 1,594 | https://github.com/huggingface/diffusers/issues/1594 | https://api.github.com/repos/huggingface/diffusers/issues/1594 | Img2Img for Video Generation | ### Model/Pipeline/Scheduler description
The proposed new community pipeline is stemmed from interpolate_stable_diffusion.py. Unlike interpolate_stable_diffusion, the proposed pipeline is interpolating between an initial image supplied by the user and an image generated by the user prompt.
At the following githu... | closed | completed | false | 5 | [
"New pipeline/model"
] | [] | 2022-12-07T15:44:01Z | 2024-11-21T22:58:20Z | 2024-11-21T22:58:20Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | MinSong2 | 13,410,808 | MDQ6VXNlcjEzNDEwODA4 | User | false |
huggingface/diffusers | 1,482,296,117 | I_kwDOHa8MBc5YWgs1 | 1,595 | https://github.com/huggingface/diffusers/issues/1595 | https://api.github.com/repos/huggingface/diffusers/issues/1595 | Natively support loading CKPT files | Hi, I'm one of the developers of [InvokeAI](https://github.com/invoke-ai/InvokeAI), a Stable Diffusion web UI targetting professionals and content creation industry use cases.
We are migrating from a CompVis to a Diffusers backend, and have recently determined that Diffusers cannot actually load ckpt models natively... | closed | completed | false | 31 | [
"stale"
] | [] | 2022-12-07T16:15:13Z | 2023-02-09T09:40:49Z | 2023-02-07T15:03:46Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | damian0815 | 144,366 | MDQ6VXNlcjE0NDM2Ng== | User | false |
huggingface/diffusers | 1,483,268,067 | I_kwDOHa8MBc5YaN_j | 1,599 | https://github.com/huggingface/diffusers/issues/1599 | https://api.github.com/repos/huggingface/diffusers/issues/1599 | [Community Pipelines] lpw_stable_diffusion no longer has attention slicing methods | Just wondering why the attention slicing was removed from LPW pipe? Also secondarily, wondering what's up with just changing `init_image` to `image`? Is there a specific reason for breaking everyone's implementations? 😂 Easy fix, just surprised. | closed | completed | false | 9 | [] | [] | 2022-12-08T01:08:04Z | 2022-12-11T19:31:04Z | 2022-12-11T19:31:00Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | WASasquatch | 1,151,589 | MDQ6VXNlcjExNTE1ODk= | User | false |
huggingface/diffusers | 1,483,357,752 | I_kwDOHa8MBc5Yaj44 | 1,600 | https://github.com/huggingface/diffusers/issues/1600 | https://api.github.com/repos/huggingface/diffusers/issues/1600 | Runtime error in image2image addnoise | ### Describe the bug
Starting this morning a scheduler in the image2image pipeline began to throw an error in addnoise.
I should add that this code was working fine yesterday and is entirely unaltered and that other, also previously working unaltered image2image projects all repeat the same error.
It is on li... | closed | completed | false | 3 | [
"bug"
] | [] | 2022-12-08T01:54:20Z | 2022-12-09T17:32:33Z | 2022-12-09T17:32:33Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | TheOneTrueGuy | 3,240,071 | MDQ6VXNlcjMyNDAwNzE= | User | false |
huggingface/diffusers | 1,483,912,503 | I_kwDOHa8MBc5YcrU3 | 1,602 | https://github.com/huggingface/diffusers/issues/1602 | https://api.github.com/repos/huggingface/diffusers/issues/1602 | DDIM Sampler on Stable-Diffusion do not work well with CFG guidance scale large than 6~7 | ### Describe the bug
I have noticed that the SD v1.4 or v1.5 works poorly if I swap the scheduler to DDIM and have guidance scale larger than 7.
This behavior does not seems to be obvious on other sampler such as the default PNDM scheduler though. At first I suspect it is related to the "train-test-mismatch" mentio... | closed | completed | false | 5 | [
"bug"
] | [] | 2022-12-08T07:22:40Z | 2022-12-13T02:39:46Z | 2022-12-13T02:39:46Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | Randolph-zeng | 11,933,185 | MDQ6VXNlcjExOTMzMTg1 | User | false |
huggingface/diffusers | 1,484,391,791 | I_kwDOHa8MBc5YegVv | 1,604 | https://github.com/huggingface/diffusers/issues/1604 | https://api.github.com/repos/huggingface/diffusers/issues/1604 | lpw_stable_diffusion.py broken when using diffusers==0.9.0 | ### Describe the bug
When initializing a diffusers pipeline, the following error occurs. Possibly related to the following [updates ](https://github.com/huggingface/diffusers/pull/1570/files)recently?
```
File ".cache/huggingface/modules/diffusers_modules/git/lpw_stable_diffusion.py", line 755, in __call__
... | closed | completed | false | 6 | [
"bug"
] | [] | 2022-12-08T11:11:36Z | 2022-12-12T15:26:05Z | 2022-12-09T07:30:28Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | kovtcharov | 4,722,733 | MDQ6VXNlcjQ3MjI3MzM= | User | false |
huggingface/diffusers | 1,484,524,272 | I_kwDOHa8MBc5YfArw | 1,606 | https://github.com/huggingface/diffusers/issues/1606 | https://api.github.com/repos/huggingface/diffusers/issues/1606 | Question about thread safety | We are thinking if we can reuse the single pipe, created on startup, between multiple threads from ThreadPoolExecutor. Is it thread safe? | closed | completed | false | 7 | [
"stale"
] | [] | 2022-12-08T12:22:33Z | 2023-07-05T03:14:14Z | 2023-01-18T15:03:41Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | miroljub1995 | 46,632,388 | MDQ6VXNlcjQ2NjMyMzg4 | User | false |
huggingface/diffusers | 1,484,695,704 | I_kwDOHa8MBc5YfqiY | 1,607 | https://github.com/huggingface/diffusers/issues/1607 | https://api.github.com/repos/huggingface/diffusers/issues/1607 | [Community Pipelines] Why does diffusers pipelines get's updated in background? Without permission? | ### Describe the bug
Now, i have working environment where i see community pipelines get's updated without any explicit updates. And now all production code has stopped working.
I am getting :: AttributeError: 'StableDiffusionLongPromptWeightingPipeline' object has no attribute 'vae_scale_factor' which wasn't th... | closed | completed | false | 19 | [
"bug",
"stale"
] | [] | 2022-12-08T13:55:16Z | 2023-01-23T15:03:49Z | 2023-01-23T15:03:49Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | adhikjoshi | 11,740,719 | MDQ6VXNlcjExNzQwNzE5 | User | false |
huggingface/diffusers | 1,484,774,906 | I_kwDOHa8MBc5Yf936 | 1,608 | https://github.com/huggingface/diffusers/issues/1608 | https://api.github.com/repos/huggingface/diffusers/issues/1608 | Training stops when reaching 500 steps on T4 | Im using `AWS T4 GPU`, and training always stops when reaching steps `500`
Error im getting:
```
Traceback (most recent call last):
File "diffusers/examples/dreambooth/train_dreambooth.py", line 719, in <module>
main(args)
File "diffusers/examples/dreambooth/train_dreambooth.py", line 638, in main
... | closed | completed | false | 9 | [] | [] | 2022-12-08T14:34:10Z | 2022-12-13T15:07:08Z | 2022-12-13T15:07:07Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | ariand9 | 6,691,197 | MDQ6VXNlcjY2OTExOTc= | User | false |
huggingface/diffusers | 1,484,786,772 | I_kwDOHa8MBc5YgAxU | 1,609 | https://github.com/huggingface/diffusers/issues/1609 | https://api.github.com/repos/huggingface/diffusers/issues/1609 | How to retrain a image-condition model on my own dataset? | @kig @kashif @srush @StoneCypher @keturn
Hello everyone!!!
I have facused on this repo long time and had difficulted in retraining a image-condition model on my own dataset.
I want to etrain a image-condition model on my own dataset ,but i can not find a config suit for my task.
Maybe you can share a example conf... | closed | completed | false | 7 | [
"stale"
] | [] | 2022-12-08T14:40:40Z | 2023-11-14T08:06:41Z | 2023-02-14T15:04:16Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | pokameng | 47,957,015 | MDQ6VXNlcjQ3OTU3MDE1 | User | false |
huggingface/diffusers | 1,484,879,274 | I_kwDOHa8MBc5YgXWq | 1,611 | https://github.com/huggingface/diffusers/issues/1611 | https://api.github.com/repos/huggingface/diffusers/issues/1611 | StableDiffusionMegaPipeline in v0.6.0 stopped working with runwayml/stable-diffusion-v1-5 | ### Describe the bug
There following code stopped working since ~3 days ago:
```
DiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
custom_pipeline="stable_diffusion_mega",
torch_dtype=torch.float16,
revision="fp16",
scheduler=StableD... | closed | completed | false | 2 | [
"bug"
] | [] | 2022-12-08T15:27:39Z | 2022-12-13T19:19:25Z | 2022-12-13T19:19:25Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | NiMaZi | 19,431,549 | MDQ6VXNlcjE5NDMxNTQ5 | User | false |
huggingface/diffusers | 1,485,127,604 | I_kwDOHa8MBc5YhT-0 | 1,614 | https://github.com/huggingface/diffusers/issues/1614 | https://api.github.com/repos/huggingface/diffusers/issues/1614 | Blank black image output with stable diffusion 2.1 using autocast | ### Describe the bug
Using stable diffusion pipeline with torch.autocast and the stabilityai/stable-diffusion-2-1 model, the images generate are all blank black images.
### Reproduction
```python
import torch
from diffusers import StableDiffusionPipeline, EulerDiscreteScheduler
model_id = "stabilityai/sta... | closed | completed | false | 6 | [
"bug",
"stale"
] | [] | 2022-12-08T17:54:44Z | 2023-02-19T15:03:04Z | 2023-02-19T15:03:04Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | fralumz | 11,469,871 | MDQ6VXNlcjExNDY5ODcx | User | false |
huggingface/diffusers | 1,485,326,379 | I_kwDOHa8MBc5YiEgr | 1,616 | https://github.com/huggingface/diffusers/issues/1616 | https://api.github.com/repos/huggingface/diffusers/issues/1616 | [Community Pipelines] LPW Pipeline on Diffusers 0.8.0 | ### Describe the bug
Why is current changes to diffusers effecting past versions? I suddenly can't use LPW on 0.8.0.
```
[/usr/local/lib/python3.8/dist-packages/diffusers/pipeline_utils.py](https://localhost:8080/#) in from_pretrained(cls, pretrained_model_name_or_path, **kwargs)
657 elif len(miss... | closed | completed | false | 6 | [
"bug",
"stale"
] | [] | 2022-12-08T20:07:01Z | 2023-02-14T15:04:13Z | 2023-02-14T15:04:13Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | WASasquatch | 1,151,589 | MDQ6VXNlcjExNTE1ODk= | User | false |
huggingface/diffusers | 1,485,396,188 | I_kwDOHa8MBc5YiVjc | 1,617 | https://github.com/huggingface/diffusers/issues/1617 | https://api.github.com/repos/huggingface/diffusers/issues/1617 | train_dreambooth.py: error: the following arguments are required: | I'm trying to fit train SD 1.5 with DreamBooth with prior preservation following the steps from [https://github.com/huggingface/diffusers/tree/main/examples/dreambooth]
But i'm receiving an error:
`train_dreambooth.py: error: the following arguments are required: --pretrained_model_name_or_path, --instance_data_di... | closed | completed | false | 8 | [
"training"
] | [] | 2022-12-08T21:01:16Z | 2022-12-12T18:12:08Z | 2022-12-12T18:12:07Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | mcventur | 8,007,246 | MDQ6VXNlcjgwMDcyNDY= | User | false |
huggingface/diffusers | 1,485,496,176 | I_kwDOHa8MBc5Yit9w | 1,619 | https://github.com/huggingface/diffusers/issues/1619 | https://api.github.com/repos/huggingface/diffusers/issues/1619 | Adding additional input channels to model after intialization / Converting text2img to mask inpainting to resume training | Have scoured the docs for an answer to this, to no avail. Is it possible to add additional input channels to a model after initializing it using `.from_pretrained`.
For example (taken from your Dreambooth example):
```
unet = UNet2DConditionModel.from_pretrained(
args.pretrained_model_name_or_path,
... | closed | completed | false | 19 | [
"stale"
] | [] | 2022-12-08T22:13:50Z | 2024-09-21T15:58:39Z | 2023-01-17T15:03:54Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | BenjaminIrwin | 59,652,119 | MDQ6VXNlcjU5NjUyMTE5 | User | false |
huggingface/diffusers | 1,486,669,122 | I_kwDOHa8MBc5YnMVC | 1,626 | https://github.com/huggingface/diffusers/issues/1626 | https://api.github.com/repos/huggingface/diffusers/issues/1626 | How to use a specific model from huggingface or Google Drive to train the images / How to load model locally? | I would like to train subjects in a specific model, how can I use it here instead of the original stable diffusion model 1.4, 1.5, 2.0, ecc | closed | completed | false | 3 | [
"stale"
] | [] | 2022-12-09T12:24:43Z | 2023-01-17T15:03:50Z | 2023-01-17T15:03:50Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | dvelm | 112,973,030 | U_kgDOBrvU5g | User | false |
huggingface/diffusers | 1,486,708,254 | I_kwDOHa8MBc5YnV4e | 1,628 | https://github.com/huggingface/diffusers/issues/1628 | https://api.github.com/repos/huggingface/diffusers/issues/1628 | Add more tests to ensure consistency for different scenarios | We have some important features in `diffusers` that are currently left untested which needs to be solved. Here a list of things that should be solved:
- [ ] 1. Add tests to make sure batching works
- [ ] 2. Add tests to make sure every PIL image can also be passed as a PyTorch tensor
- [ ] 3. Add a signature test ... | closed | completed | false | 2 | [
"stale"
] | [] | 2022-12-09T12:51:40Z | 2023-02-08T15:07:17Z | 2023-02-08T15:07:17Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | patrickvonplaten | 23,423,619 | MDQ6VXNlcjIzNDIzNjE5 | User | false |
huggingface/diffusers | 1,486,754,437 | I_kwDOHa8MBc5YnhKF | 1,629 | https://github.com/huggingface/diffusers/issues/1629 | https://api.github.com/repos/huggingface/diffusers/issues/1629 | [Tests] `AltDiffusionPipelineFastTests` not fully reproducible | The fast tests for AltDiffusion currently use `CLIPTextModel` instead of `RobertaSeriesModel` because the latter does not produce the same outputs between two consecutive runs, breaking common tests like save-loading and dict-tuple outputs comparison.
We should figure out the source of non-determinism and use Robert... | closed | completed | false | 1 | [
"stale"
] | [
"patrickvonplaten"
] | 2022-12-09T13:21:31Z | 2023-02-14T15:04:10Z | 2023-02-14T15:04:10Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | anton-l | 26,864,830 | MDQ6VXNlcjI2ODY0ODMw | User | false |
huggingface/diffusers | 1,334,705,540 | I_kwDOHa8MBc5Pjf2E | 163 | https://github.com/huggingface/diffusers/issues/163 | https://api.github.com/repos/huggingface/diffusers/issues/163 | Latent diffusion model | Hi,
Is there a way to make the latent diffusion model able to get the context from any modality (that would be user defined) with a training script associated to this ? (modality as features, text or basically encoder that outpus some latent vector of an other modality )
Let me know if its not clear :)
Thank you !
| closed | completed | false | 9 | [
"stale"
] | [] | 2022-08-10T14:20:12Z | 2022-11-05T15:03:17Z | 2022-11-05T15:03:17Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | ethancohen123 | 55,588,944 | MDQ6VXNlcjU1NTg4OTQ0 | User | false |
huggingface/diffusers | 1,486,769,612 | I_kwDOHa8MBc5Ynk3M | 1,630 | https://github.com/huggingface/diffusers/issues/1630 | https://api.github.com/repos/huggingface/diffusers/issues/1630 | [Tests] Inconsistent outputs for common fast tests on MPS | There are three (so far) pipelines that don't produce the same outputs between two consecutive runs on MPS, despite having a warmup step in the common fast tests:
https://github.com/huggingface/diffusers/blob/cd91fc06fe9513864fca6a57953ca85a7ae7836e/tests/test_pipelines_common.py#L67-L73
Needs further investigati... | closed | completed | false | 4 | [
"stale"
] | [
"pcuenca",
"patrickvonplaten",
"anton-l"
] | 2022-12-09T13:31:49Z | 2023-02-14T15:04:09Z | 2023-02-14T15:04:09Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | anton-l | 26,864,830 | MDQ6VXNlcjI2ODY0ODMw | User | false |
huggingface/diffusers | 1,486,783,925 | I_kwDOHa8MBc5YnoW1 | 1,631 | https://github.com/huggingface/diffusers/issues/1631 | https://api.github.com/repos/huggingface/diffusers/issues/1631 | [Tests] Non-deterministic depth estimator in `StableDiffusiondepth2imgPipelineFastTests` | `DPTForDepthEstimation` might be causing less reproducible results between two consecutive runs than other transformers models, which motivated overriding the tolerances of common fast tests, e.g. `3e-5` here:
https://github.com/huggingface/diffusers/blob/cd91fc06fe9513864fca6a57953ca85a7ae7836e/tests/pipelines/stab... | closed | completed | false | 4 | [
"stale"
] | [] | 2022-12-09T13:41:53Z | 2023-02-14T15:04:07Z | 2023-02-14T15:04:07Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | anton-l | 26,864,830 | MDQ6VXNlcjI2ODY0ODMw | User | false |
huggingface/diffusers | 1,486,788,259 | I_kwDOHa8MBc5Ynpaj | 1,632 | https://github.com/huggingface/diffusers/issues/1632 | https://api.github.com/repos/huggingface/diffusers/issues/1632 | UNet1DModel Input Tensor shape | Hello,
I am trying to use a custom [Unet1DModel](https://github.com/huggingface/diffusers/blob/3faf204c496ffed9ad48862874db7031bbe2674e/src/diffusers/models/unet_1d.py). I got it running with the[ code that is used for testing](https://github.com/huggingface/diffusers/blob/3faf204c496ffed9ad48862874db7031bbe2674e/te... | closed | completed | false | 4 | [] | [] | 2022-12-09T13:44:56Z | 2023-09-23T17:35:04Z | 2022-12-13T09:41:36Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | SvenGroen | 38,216,862 | MDQ6VXNlcjM4MjE2ODYy | User | false |
huggingface/diffusers | 1,486,791,997 | I_kwDOHa8MBc5YnqU9 | 1,633 | https://github.com/huggingface/diffusers/issues/1633 | https://api.github.com/repos/huggingface/diffusers/issues/1633 | Support optionally using karras sigmas in the k_diffusion wrapper pipeline | **Is your feature request related to a problem? Please describe.**
K-Diffusion enables users to generate sigmas with `k_diffusion.sampling.get_sigmas_karras`. `diffusers` currently do not support using those to generate the sigmas with its [k_diffusion wrapper](https://github.com/huggingface/diffusers/blob/main/exampl... | closed | completed | false | 8 | [
"wip"
] | [] | 2022-12-09T13:47:32Z | 2024-11-21T15:29:38Z | 2024-11-21T15:29:38Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | apolinario | 788,417 | MDQ6VXNlcjc4ODQxNw== | User | false |
huggingface/diffusers | 1,486,794,629 | I_kwDOHa8MBc5Ynq-F | 1,634 | https://github.com/huggingface/diffusers/issues/1634 | https://api.github.com/repos/huggingface/diffusers/issues/1634 | [MPS] Some of the `DPTForDepthEstimation` ops are not supported on MPS | To reproduce, remove the mps skips in `StableDiffusiondepth2imgPipelineFastTests`:
https://github.com/huggingface/diffusers/blob/cd91fc06fe9513864fca6a57953ca85a7ae7836e/tests/pipelines/stable_diffusion_2/test_stable_diffusion_depth.py#L154
This is probably solvable on the `transformers` side, but opening an issu... | closed | completed | false | 4 | [
"bug"
] | [
"pcuenca"
] | 2022-12-09T13:49:22Z | 2023-02-15T15:03:34Z | 2023-02-15T15:03:34Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | anton-l | 26,864,830 | MDQ6VXNlcjI2ODY0ODMw | User | false |
huggingface/diffusers | 1,486,893,792 | I_kwDOHa8MBc5YoDLg | 1,635 | https://github.com/huggingface/diffusers/issues/1635 | https://api.github.com/repos/huggingface/diffusers/issues/1635 | [Quick poll] Give your opinion on the future of the Hugging Face Open Source ecosystem! | Thanks to all of you, Diffusers just passed 8.5k :star2: !
Since the last survey, a lot has happened: the [diffusers](https://github.com/huggingface/diffusers), [evaluate](https://github.com/huggingface/evaluate) and [skops](https://github.com/skops-dev/skops) libraries were born. `timm` joined the Hugging Face ecos... | closed | completed | false | 1 | [
"stale"
] | [] | 2022-12-09T14:53:51Z | 2023-01-17T15:03:46Z | 2023-01-17T15:03:46Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | LysandreJik | 30,755,778 | MDQ6VXNlcjMwNzU1Nzc4 | User | false |
huggingface/diffusers | 1,486,911,744 | I_kwDOHa8MBc5YoHkA | 1,636 | https://github.com/huggingface/diffusers/issues/1636 | https://api.github.com/repos/huggingface/diffusers/issues/1636 | Dynamic thresholding | Could be important for increased quality or robustness to "fried" areas. | closed | completed | false | 9 | [
"wip"
] | [] | 2022-12-09T15:05:15Z | 2023-08-02T13:35:31Z | 2023-04-27T01:35:01Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | pcuenca | 1,177,582 | MDQ6VXNlcjExNzc1ODI= | User | false |
huggingface/diffusers | 1,487,261,674 | I_kwDOHa8MBc5Ypc_q | 1,641 | https://github.com/huggingface/diffusers/issues/1641 | https://api.github.com/repos/huggingface/diffusers/issues/1641 | I want to use "max_embeddings_multiples" argument to increase over 75 tokens. | I want to use "max_embeddings_multiples" argument to increase over 75 tokens.
I have this code.
```
from diffusers import DiffusionPipeline, AutoencoderKL
import torch
import random
from IPython.display import Image
model_id = "./anything-v3.0"
vae = AutoencoderKL.from_pretrained("./anything_vae")
pipe = D... | closed | completed | false | 4 | [] | [] | 2022-12-09T18:49:58Z | 2022-12-10T14:19:34Z | 2022-12-10T14:19:34Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | NNMYYTK | 79,393,842 | MDQ6VXNlcjc5MzkzODQy | User | false |
huggingface/diffusers | 1,488,108,570 | I_kwDOHa8MBc5Ysrwa | 1,642 | https://github.com/huggingface/diffusers/issues/1642 | https://api.github.com/repos/huggingface/diffusers/issues/1642 | Forward pass of ScoreSdeVeScheduler scheduler is missing | ### Describe the bug
Hi,
Forward pass is missing for this scheduler
https://github.com/huggingface/diffusers/blob/v0.10.2/src/diffusers/schedulers/scheduling_sde_ve.py#L45
It would be great if this could be added. Could you give me some pointers how I can implement the forward pass?
thanks
### Reproduction... | closed | completed | false | 4 | [
"bug",
"stale"
] | [] | 2022-12-10T06:57:24Z | 2023-01-17T15:03:44Z | 2023-01-17T15:03:44Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | cameronalonso2 | 73,615,224 | MDQ6VXNlcjczNjE1MjI0 | User | false |
huggingface/diffusers | 1,488,373,551 | I_kwDOHa8MBc5Ytscv | 1,643 | https://github.com/huggingface/diffusers/issues/1643 | https://api.github.com/repos/huggingface/diffusers/issues/1643 | dreambooth : ModelMixin.enable_xformers_memory_efficient_attention() takes 1 positional argument but 2 were given | ### Describe the bug
I cant get xformers to run if I only run
from diffusers.utils.import_utils import is_xformers_available
if is_xformers_available():
It says its available.
The code is the train_dreambooth.py script
Thank you!
### Reproduction
vanilla train_dreambooth.py from the examples
### Logs... | closed | completed | false | 0 | [
"bug"
] | [] | 2022-12-10T10:51:18Z | 2022-12-10T12:57:21Z | 2022-12-10T12:57:21Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | UglyStupidHonest | 68,284,461 | MDQ6VXNlcjY4Mjg0NDYx | User | false |
huggingface/diffusers | 1,488,524,312 | I_kwDOHa8MBc5YuRQY | 1,646 | https://github.com/huggingface/diffusers/issues/1646 | https://api.github.com/repos/huggingface/diffusers/issues/1646 | Should we open Github Discussions? | Is it possible to enable github discussions then people can ask questions and also we can move some issues to discussions if it is a question. | closed | completed | false | 2 | [] | [] | 2022-12-10T14:15:19Z | 2022-12-11T16:17:59Z | 2022-12-11T16:17:59Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | camenduru | 54,370,274 | MDQ6VXNlcjU0MzcwMjc0 | User | false |
huggingface/diffusers | 1,488,529,089 | I_kwDOHa8MBc5YuSbB | 1,647 | https://github.com/huggingface/diffusers/issues/1647 | https://api.github.com/repos/huggingface/diffusers/issues/1647 | [Question][Solved] How can I use color correction .pt in diffusers | https://huggingface.co/Linaqruf/anything-v3.0/blob/main/Anything-V3.0.vae.pt
this is for color correction how can I use this with diffusers | closed | completed | false | 6 | [] | [] | 2022-12-10T14:21:40Z | 2023-02-04T16:28:32Z | 2022-12-22T23:36:28Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | camenduru | 54,370,274 | MDQ6VXNlcjU0MzcwMjc0 | User | false |
huggingface/diffusers | 1,488,663,033 | I_kwDOHa8MBc5YuzH5 | 1,651 | https://github.com/huggingface/diffusers/issues/1651 | https://api.github.com/repos/huggingface/diffusers/issues/1651 | Dreambooth training deadlock on GCP instances with pre-built PyTorch images | I am trying to reproduce the 'Dog toy example' in https://github.com/huggingface/diffusers/tree/main/examples/dreambooth but I am getting stuck at https://github.com/huggingface/diffusers/blob/main/examples/dreambooth/train_dreambooth.py#L675 forever. Do you know how to fix it?
I have followed through the instructi... | closed | completed | false | 12 | [
"stale",
"training"
] | [
"pcuenca"
] | 2022-12-10T16:44:08Z | 2023-02-10T17:45:26Z | 2023-02-10T17:45:26Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | zihao12 | 34,700,423 | MDQ6VXNlcjM0NzAwNDIz | User | false |
huggingface/diffusers | 1,489,001,873 | I_kwDOHa8MBc5YwF2R | 1,654 | https://github.com/huggingface/diffusers/issues/1654 | https://api.github.com/repos/huggingface/diffusers/issues/1654 | [Trivial Fix] Flag use_ema in train_unconditional.py should not have default value | ## Problem
The definition of the flag `use_ema` wrongly includes a default value of `True`:
https://github.com/huggingface/diffusers/blob/fc94c60c8373862c509e388f3f4065d98cedf589/examples/unconditional_image_generation/train_unconditional.py#L156-L161
Hence, when the flag is not passed via CLI, **it is still evaluat... | closed | completed | false | 2 | [
"stale"
] | [
"anton-l"
] | 2022-12-10T21:41:31Z | 2023-02-14T15:04:04Z | 2023-02-14T15:04:04Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | nkirschi | 22,997,656 | MDQ6VXNlcjIyOTk3NjU2 | User | false |
huggingface/diffusers | 1,489,317,371 | I_kwDOHa8MBc5YxS37 | 1,655 | https://github.com/huggingface/diffusers/issues/1655 | https://api.github.com/repos/huggingface/diffusers/issues/1655 | StableDiffusionImageVariationPipeline not initializing | ### Describe the bug
When I try to initialize pipe I get this error:
`requests.exceptions.HTTPError: 404 Client Error: Not Found for url: https://huggingface.co/lambdalabs/sd-image-variations-diffusers/resolve/v2.0/model_index.json`
There is a problem with lambdalabs repo, please check. This issue can also be fo... | closed | completed | false | 3 | [
"bug",
"stale"
] | [] | 2022-12-11T02:48:53Z | 2023-01-18T15:03:31Z | 2023-01-18T15:03:31Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | generalsvr | 32,482,126 | MDQ6VXNlcjMyNDgyMTI2 | User | false |
huggingface/diffusers | 1,489,559,803 | I_kwDOHa8MBc5YyOD7 | 1,656 | https://github.com/huggingface/diffusers/issues/1656 | https://api.github.com/repos/huggingface/diffusers/issues/1656 | Question about noise schedule parametrization & samplers | Thanks for all the work first of all, I appreciate that we have all these pretrained models freely available open-source.
I'm a bit confused by the relationship between sampler implementations and model checkpoints. If someone could clarify the following, that'd be much appreciated (and perhaps some of these can be... | closed | completed | false | 3 | [
"stale"
] | [] | 2022-12-11T06:52:46Z | 2023-01-18T15:03:30Z | 2023-01-18T15:03:29Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | jaywhang | 2,825,858 | MDQ6VXNlcjI4MjU4NTg= | User | false |
huggingface/diffusers | 1,490,101,946 | I_kwDOHa8MBc5Y0Sa6 | 1,657 | https://github.com/huggingface/diffusers/issues/1657 | https://api.github.com/repos/huggingface/diffusers/issues/1657 | Any kaggle notebook for dreambooth examples? | anyone has setup a kaggle notebook that we can load our images train and generate outputs?
I am talking about this one : https://github.com/huggingface/diffusers/tree/main/examples/dreambooth | closed | completed | false | 2 | [
"stale",
"training"
] | [] | 2022-12-11T15:09:32Z | 2023-01-20T15:03:13Z | 2023-01-20T15:03:13Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | FurkanGozukara | 19,240,467 | MDQ6VXNlcjE5MjQwNDY3 | User | false |
huggingface/diffusers | 1,490,127,329 | I_kwDOHa8MBc5Y0Ynh | 1,658 | https://github.com/huggingface/diffusers/issues/1658 | https://api.github.com/repos/huggingface/diffusers/issues/1658 | Error When Training Dreambooth after Saving Weights: RuntimeError: Inference tensors cannot be saved for backward. | ### Describe the bug
Using my custom version of dreambooth for Auto1111, I've started encountering a new issue after saving a checkpoint that I'm like 95% sure I didn't cause...
https://github.com/d8ahazard/sd_dreambooth_extension/blob/main/dreambooth/train_dreambooth.py
I have a feeling this is similar to or re... | closed | completed | false | 4 | [
"bug",
"stale"
] | [] | 2022-12-11T15:32:28Z | 2023-05-18T10:07:14Z | 2023-02-07T15:03:38Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | d8ahazard | 1,633,844 | MDQ6VXNlcjE2MzM4NDQ= | User | false |
huggingface/diffusers | 1,490,268,114 | I_kwDOHa8MBc5Y06_S | 1,659 | https://github.com/huggingface/diffusers/issues/1659 | https://api.github.com/repos/huggingface/diffusers/issues/1659 | (WSL2) Pipelines loaded with `torch_dtype=torch.float16` cannot run with `cpu` device. It is not recommended to move them to `cpu` as running them will fail. Please make sure to use an accelerator to run the pipeline in inference, due to the lack of support for`float16` operations on this device in PyTorch. Please, rem... | > Pipelines loaded with `torch_dtype=torch.float16` cannot run with `cpu` device. It is not recommended to move them to `cpu` as running them will fail. Please make sure to use an accelerator to run the pipeline in inference, due to the lack of support for`float16` operations on this device in PyTorch. Please, remove t... | closed | completed | false | 7 | [
"stale"
] | [] | 2022-12-11T17:45:33Z | 2023-09-28T15:00:42Z | 2023-01-20T15:03:11Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | darkanubis0100 | 22,696,317 | MDQ6VXNlcjIyNjk2MzE3 | User | false |
huggingface/diffusers | 1,336,288,646 | I_kwDOHa8MBc5PpiWG | 166 | https://github.com/huggingface/diffusers/issues/166 | https://api.github.com/repos/huggingface/diffusers/issues/166 | Frozen parameters in GaussianFourierProjection | Hi, just a beginner with diffusion models and have been using your implementations as reference. I have a question about [this class](https://github.com/huggingface/diffusers/blob/3a32b8c916beced18f721f1c95f9cbd6c4c4aacb/src/diffusers/models/embeddings.py#L94)
Why is `requires_grad` set to false in the weight parame... | closed | completed | false | 5 | [] | [] | 2022-08-11T17:51:08Z | 2022-08-31T08:10:30Z | 2022-08-31T08:10:30Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | vvvm23 | 44,398,246 | MDQ6VXNlcjQ0Mzk4MjQ2 | User | false |
huggingface/diffusers | 1,490,433,372 | I_kwDOHa8MBc5Y1jVc | 1,660 | https://github.com/huggingface/diffusers/issues/1660 | https://api.github.com/repos/huggingface/diffusers/issues/1660 | Add back **kwargs to pipelines for custom development | **Is your feature request related to a problem? Please describe.**
Removing `**kwargs` has no clear use other than making development and customization of diffusers more a chore.
Being able to pass anything to a pipe, and utilize it within diffusers API was nothing but beneficial, and I don't see why it was remove... | closed | completed | false | 8 | [] | [] | 2022-12-11T20:54:03Z | 2022-12-15T20:57:30Z | 2022-12-12T10:16:36Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | WASasquatch | 1,151,589 | MDQ6VXNlcjExNTE1ODk= | User | false |
huggingface/diffusers | 1,490,526,211 | I_kwDOHa8MBc5Y16AD | 1,662 | https://github.com/huggingface/diffusers/issues/1662 | https://api.github.com/repos/huggingface/diffusers/issues/1662 | FlaxStableDiffusionPipeline fails to load StableDiffusion 2.0 | ### Describe the bug
I can't load the SD 2.0 model using the FlaxStableDiffusionPipeline. The SD 1.5 models seems to load ok.
While on that topic, `FlaxStableDiffusionPipeline` could do with better documentation:
- Looking at the tests, it seems that flax models need to use revision="bf16" and torch_dtype=jnp.bflo... | closed | completed | false | 4 | [
"bug",
"stale"
] | [] | 2022-12-11T22:40:12Z | 2023-03-01T15:09:37Z | 2023-01-20T15:03:09Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | antoche | 1,627,384 | MDQ6VXNlcjE2MjczODQ= | User | false |
huggingface/diffusers | 1,491,563,594 | I_kwDOHa8MBc5Y53RK | 1,663 | https://github.com/huggingface/diffusers/issues/1663 | https://api.github.com/repos/huggingface/diffusers/issues/1663 | Store cached models in location compatible with `huggingface-cli` | Reference: https://github.com/huggingface/huggingface_hub/issues/1259
`huggingface-cli scan-cache` doesn't see cached diffusers models. Are there any drawbacks to changing the cache folder? | closed | completed | false | 21 | [
"wip"
] | [
"pcuenca",
"patrickvonplaten"
] | 2022-12-12T10:49:28Z | 2023-07-19T10:40:19Z | 2023-02-28T19:01:04Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | pcuenca | 1,177,582 | MDQ6VXNlcjExNzc1ODI= | User | false |
huggingface/diffusers | 1,492,081,614 | I_kwDOHa8MBc5Y71vO | 1,666 | https://github.com/huggingface/diffusers/issues/1666 | https://api.github.com/repos/huggingface/diffusers/issues/1666 | Minor inconsistency for scheduler in the example for stable-diffusion-2-1 | In [the example for `stabilityai/stable-diffusion-2-1`](https://huggingface.co/stabilityai/stable-diffusion-2-1#examples), you mention that:
> if you don't swap the scheduler it will run with the default DDIM,
> **in this example we are swapping it to EulerDiscreteScheduler**
This also appears in a comment in Py... | closed | completed | false | 2 | [] | [] | 2022-12-12T14:59:40Z | 2022-12-12T17:13:17Z | 2022-12-12T17:13:17Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | woctezuma | 570,669 | MDQ6VXNlcjU3MDY2OQ== | User | false |
huggingface/diffusers | 1,492,678,434 | I_kwDOHa8MBc5Y-Hci | 1,672 | https://github.com/huggingface/diffusers/issues/1672 | https://api.github.com/repos/huggingface/diffusers/issues/1672 | Images are the wrong size (256px for stable diffusion 1.5) after using ckpt to diffusers script | ### Describe the bug
Hello, I was testing [the conversion script for ckpts to diffusers](https://raw.githubusercontent.com/huggingface/diffusers/039958eae55ff0700cfb42a7e72739575ab341f1/scripts/convert_original_stable_diffusion_to_diffusers.py), when I realized all the images that generated by the converted models are... | closed | completed | false | 4 | [
"bug",
"stale"
] | [] | 2022-12-12T20:17:57Z | 2023-01-22T15:02:56Z | 2023-01-22T15:02:56Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | JohnnyRacer | 77,214,388 | MDQ6VXNlcjc3MjE0Mzg4 | User | false |
huggingface/diffusers | 1,492,682,273 | I_kwDOHa8MBc5Y-IYh | 1,673 | https://github.com/huggingface/diffusers/issues/1673 | https://api.github.com/repos/huggingface/diffusers/issues/1673 | dreambooth (minor): command line handling: accelerate logger used before being initialized | ### Describe the bug
As I was working on another feature, I realized that parse_args uses `logger`. But `logger` isn't ready until after parse_args returns, as it depends on a configured `Accelerator`.
### Reproduction
Passing `--class_data_dir` but not `--with_prior_preservation` tries to write a warning, b... | closed | completed | false | 1 | [
"bug"
] | [
"pcuenca"
] | 2022-12-12T20:20:08Z | 2022-12-13T21:02:35Z | 2022-12-13T21:01:49Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | timh | 139,355 | MDQ6VXNlcjEzOTM1NQ== | User | false |
huggingface/diffusers | 1,492,764,888 | I_kwDOHa8MBc5Y-cjY | 1,674 | https://github.com/huggingface/diffusers/issues/1674 | https://api.github.com/repos/huggingface/diffusers/issues/1674 | [Custom Models] Can't load dreambooth models with new scheduling config method and DPMSolver | ### Describe the bug
I can't load DPMSolver, for example, with the new method to define schedulers from config, cause apparently, most these dreambooth models don't have a config for the schedulers. What do we do in this instance then?
```OSError: nitrosocke/redshift-diffusion does not appear to have a file name... | closed | completed | false | 7 | [
"bug"
] | [
"pcuenca"
] | 2022-12-12T21:04:08Z | 2022-12-14T23:00:26Z | 2022-12-14T23:00:26Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | WASasquatch | 1,151,589 | MDQ6VXNlcjExNTE1ODk= | User | false |
huggingface/diffusers | 1,492,817,138 | I_kwDOHa8MBc5Y-pTy | 1,675 | https://github.com/huggingface/diffusers/issues/1675 | https://api.github.com/repos/huggingface/diffusers/issues/1675 | convert_original_stable_diffusion_to_diffusers.py error state dict | convert_original_stable_diffusion_to_diffusers.py
Im trying to convert some ckpt to diffusers but im getting this error:
Traceback (most recent call last):
File "H:\stable\convert_original_stable_diffusion_to_diffusers.py", line 864, in <module>
checkpoint = checkpoint["state_dict"]
KeyError: 'state_di... | closed | completed | false | 4 | [
"stale"
] | [] | 2022-12-12T21:40:21Z | 2023-01-23T15:03:42Z | 2023-01-23T15:03:42Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | elen07zz | 105,580,965 | U_kgDOBksJpQ | User | false |
huggingface/diffusers | 1,493,033,154 | I_kwDOHa8MBc5Y_eDC | 1,676 | https://github.com/huggingface/diffusers/issues/1676 | https://api.github.com/repos/huggingface/diffusers/issues/1676 | Strange behaviour and issues with SD 2.1 on multi-gpu (4090) and xformers attention | ### Describe the bug
Hello!
Today I tried to setup SD 2.1 but have some trouble with more than one issue.
```
pipe = StableDiffusionPipeline.from_pretrained(
"stabilityai/stable-diffusion-2-1",
torch_dtype=torch.float16,
).to("cuda") # or cuda:0
```
together with
`pipe.enable_xformers_memory_e... | closed | completed | false | 7 | [
"bug",
"stale"
] | [
"pcuenca"
] | 2022-12-12T23:52:31Z | 2023-02-01T15:03:20Z | 2023-02-01T15:03:20Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | Marcophono2 | 22,599,855 | MDQ6VXNlcjIyNTk5ODU1 | User | false |
huggingface/diffusers | 1,493,437,503 | I_kwDOHa8MBc5ZBAw_ | 1,677 | https://github.com/huggingface/diffusers/issues/1677 | https://api.github.com/repos/huggingface/diffusers/issues/1677 | Call StableDiffusionPipeline object with wrong data type | ### Describe the bug
When I call pipe object, it throws Runtime Error: expected scalar type half but found float
I am not sure whether it is the dtype of latent, so I manually define the dtype of latents to fp16 but get the same error.
### Reproduction
import torch
from diffusers import StableDiffusionPipeli... | closed | completed | false | 1 | [
"bug"
] | [] | 2022-12-13T05:51:04Z | 2022-12-13T11:32:47Z | 2022-12-13T11:32:47Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | chesha1 | 49,867,392 | MDQ6VXNlcjQ5ODY3Mzky | User | false |
huggingface/diffusers | 1,493,822,702 | I_kwDOHa8MBc5ZCezu | 1,678 | https://github.com/huggingface/diffusers/issues/1678 | https://api.github.com/repos/huggingface/diffusers/issues/1678 | Best practices / example results and run settings for Dreambooth inpainting | Hi @thedarkzeno @patil-suraj
I'm trying to run the script here, https://github.com/huggingface/diffusers/tree/main/examples/research_projects/dreambooth_inpaint
(ty for this)
with similar setup as with training regular text=>image dreambooth. I'm finding that the quality of the generations for inpainting is fa... | closed | completed | false | 9 | [
"stale"
] | [] | 2022-12-13T09:35:37Z | 2023-03-04T16:36:08Z | 2023-02-13T15:03:36Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | aaronlohwy | 9,904,693 | MDQ6VXNlcjk5MDQ2OTM= | User | false |
huggingface/diffusers | 1,493,885,850 | I_kwDOHa8MBc5ZCuOa | 1,679 | https://github.com/huggingface/diffusers/issues/1679 | https://api.github.com/repos/huggingface/diffusers/issues/1679 | [Tests] Add nightly tests for popular community pipelines | To avoid breaking changes with added/deprecated functionality, as discussed in #1622 | closed | completed | false | 1 | [
"stale"
] | [
"anton-l"
] | 2022-12-13T10:07:33Z | 2023-01-21T15:02:40Z | 2023-01-21T15:02:40Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | anton-l | 26,864,830 | MDQ6VXNlcjI2ODY0ODMw | User | false |
huggingface/diffusers | 1,494,268,854 | I_kwDOHa8MBc5ZELu2 | 1,683 | https://github.com/huggingface/diffusers/issues/1683 | https://api.github.com/repos/huggingface/diffusers/issues/1683 | Opt-in telemetry (or remove telemetry) | Telemetry should be opt-in, or simply removed. | closed | completed | false | 5 | [] | [] | 2022-12-13T13:46:31Z | 2022-12-14T14:57:57Z | 2022-12-14T14:57:57Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | hlky | 106,811,348 | U_kgDOBl3P1A | User | false |
huggingface/diffusers | 1,494,273,033 | I_kwDOHa8MBc5ZEMwJ | 1,684 | https://github.com/huggingface/diffusers/issues/1684 | https://api.github.com/repos/huggingface/diffusers/issues/1684 | Getting "signal: floating point exception" error on inference call | I am trying to do an inference call on A6000 GPU but getting the following error:
```
Creating scheduler
Error: signal: floating point exception (core dumped)
```
I get the error only with A100 or A6000. I don't get this error with Quadro RTX 5000.
Although, I get CUDA out of memory with Quadro RTX 5000.
I... | closed | completed | false | 2 | [] | [] | 2022-12-13T13:49:04Z | 2022-12-21T13:55:56Z | 2022-12-21T13:55:56Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | geekyayush | 25,874,080 | MDQ6VXNlcjI1ODc0MDgw | User | false |
huggingface/diffusers | 1,494,328,617 | I_kwDOHa8MBc5ZEaUp | 1,685 | https://github.com/huggingface/diffusers/issues/1685 | https://api.github.com/repos/huggingface/diffusers/issues/1685 | It is not possible to disable telemetry | ### Describe the bug
Although the DISABLE_TELEMETRY and HF_HUB_OFFLINE env variables exist, setting them doesn't actually seem to disable telemetry.
https://github.com/huggingface/diffusers/blob/main/src/diffusers/hub_utils.py#L87
Pass is used, and /api/telemetry is prepended to the huggingface endpoint.
### Re... | closed | completed | false | 3 | [
"bug"
] | [] | 2022-12-13T14:17:22Z | 2022-12-19T11:52:51Z | 2022-12-13T15:23:25Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | w4ffl35 | 25,737,761 | MDQ6VXNlcjI1NzM3NzYx | User | false |
huggingface/diffusers | 1,494,542,692 | I_kwDOHa8MBc5ZFOlk | 1,689 | https://github.com/huggingface/diffusers/issues/1689 | https://api.github.com/repos/huggingface/diffusers/issues/1689 | `seed` missing from `StableDiffusionPipeline()` v2*? | It appears that the `seed` argument is missing from v2* of `StableDiffusionPipeline()`.
https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion.py#L397
If this is correct, I recommend editing the `generator: Optional[torch.Generator] = None` to include... | closed | completed | false | 6 | [
"stale"
] | [] | 2022-12-13T15:50:06Z | 2023-01-22T15:02:54Z | 2023-01-22T15:02:54Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | BFauber | 21,345,886 | MDQ6VXNlcjIxMzQ1ODg2 | User | false |
huggingface/diffusers | 1,494,802,554 | I_kwDOHa8MBc5ZGOB6 | 1,692 | https://github.com/huggingface/diffusers/issues/1692 | https://api.github.com/repos/huggingface/diffusers/issues/1692 | AttributeError: Can't pickle local object 'main.<locals>.<lambda>' | ### Describe the bug
Under Windows and conda, I get this error on the current diffusers version when I start examples/dreambooth/train_dreambooth.py
`AttributeError: Can't pickle local object 'main.<locals>.<lambda>'`
I solved setting the DataLoaders num_workers=0 in line 573:
https://github.com/huggingface/diffu... | closed | completed | false | 6 | [
"bug",
"stale"
] | [] | 2022-12-13T17:32:14Z | 2023-02-25T15:04:20Z | 2023-02-25T15:04:20Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | djdookie | 17,240,204 | MDQ6VXNlcjE3MjQwMjA0 | User | false |
huggingface/diffusers | 1,494,859,906 | I_kwDOHa8MBc5ZGcCC | 1,693 | https://github.com/huggingface/diffusers/issues/1693 | https://api.github.com/repos/huggingface/diffusers/issues/1693 | libbitsandbytes using older CUDA 11.6 | Hi.
When I am training a stable diffusion 2.1 768px model, I see this line in my CLI under Windows and Conda env:
CUDA SETUP: Loading binary....... libbitsandbytes_cuda116.dll
I wonder why an older CUDA verison is used here, since I have installed CUDA 11.8, torch 1.11.3 with CUDA 11.7 support (torch 1.13.0+cu117)... | closed | completed | false | 3 | [] | [] | 2022-12-13T17:58:31Z | 2022-12-19T14:12:34Z | 2022-12-19T14:12:33Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | djdookie | 17,240,204 | MDQ6VXNlcjE3MjQwMjA0 | User | false |
huggingface/diffusers | 1,495,055,992 | I_kwDOHa8MBc5ZHL54 | 1,695 | https://github.com/huggingface/diffusers/issues/1695 | https://api.github.com/repos/huggingface/diffusers/issues/1695 | Training dreambooth on multiple GPUs with deepspeed | ### Describe the bug
Hi,
I am distributing dream booth training on 8 Tesla V100 (16GB) GPUs using deep speed.
I have configured accelerate as below(output of default_config.yaml):
`command_file: null
commands: null
compute_environment: LOCAL_MACHINE
deepspeed_config:
gradient_accumulation_steps: 1
offl... | closed | completed | false | 8 | [
"bug",
"stale",
"training"
] | [
"pcuenca",
"williamberman",
"patil-suraj"
] | 2022-12-13T19:59:05Z | 2023-02-06T15:03:30Z | 2023-02-06T15:03:30Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | hamzafar | 6,842,017 | MDQ6VXNlcjY4NDIwMTc= | User | false |
huggingface/diffusers | 1,495,290,644 | I_kwDOHa8MBc5ZIFMU | 1,696 | https://github.com/huggingface/diffusers/issues/1696 | https://api.github.com/repos/huggingface/diffusers/issues/1696 | Getting Error: signal: floating point exception (core dumped) when changing the scheduler | I am trying to change the scheduler of my dreambooth model in the inference call. But I am getting the following error when I do so:
```
Error: signal: floating point exception (core dumped)
```
Here's how I am trying to do it:
```
newScheduler = DPMSolverMultistepScheduler.from_config(pipeline.scheduler.conf... | closed | completed | false | 2 | [
"stale"
] | [] | 2022-12-13T22:57:16Z | 2023-01-22T15:02:52Z | 2023-01-22T15:02:52Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | geekyayush | 25,874,080 | MDQ6VXNlcjI1ODc0MDgw | User | false |
huggingface/diffusers | 1,495,486,593 | I_kwDOHa8MBc5ZI1CB | 1,698 | https://github.com/huggingface/diffusers/issues/1698 | https://api.github.com/repos/huggingface/diffusers/issues/1698 | How to check a checkpoints output quality? | If I use --checkpointing_steps for dreambooth training, is there a way to check the output/quality of those checkpoints?
Can I automatically create sample images at each checkpoint?
Since the description says "These checkpoints are only suitable for resuming training using --resume_from_checkpoint", is there a way to... | closed | completed | false | 6 | [] | [
"pcuenca"
] | 2022-12-14T02:16:30Z | 2023-01-16T13:56:20Z | 2022-12-21T13:32:26Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | djdookie | 17,240,204 | MDQ6VXNlcjE3MjQwMjA0 | User | false |
huggingface/diffusers | 1,496,002,615 | I_kwDOHa8MBc5ZKzA3 | 1,699 | https://github.com/huggingface/diffusers/issues/1699 | https://api.github.com/repos/huggingface/diffusers/issues/1699 | Image output | Hi just a stupid question but it's driving me nuts - running in colab on defaults just testing, where do the generated images end up? or how to download? | closed | completed | false | 2 | [
"stale"
] | [] | 2022-12-14T08:02:37Z | 2023-01-22T15:02:51Z | 2023-01-22T15:02:51Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | BCusack | 10,039,059 | MDQ6VXNlcjEwMDM5MDU5 | User | false |
huggingface/diffusers | 1,337,875,451 | I_kwDOHa8MBc5Pvlv7 | 170 | https://github.com/huggingface/diffusers/issues/170 | https://api.github.com/repos/huggingface/diffusers/issues/170 | Stable Diffusion v1-3 crashes with TypeError: step() got an unexpected keyword argument 'eta' | ### Describe the bug
When attempting text-to-image synthesis, Stable Diffusion v1-3 crashes with `TypeError: step() got an unexpected keyword argument 'eta'`
EDIT: I didn't realize someone started a [thread](https://huggingface.co/CompVis/stable-diffusion-v1-3-diffusers/discussions/1) about this same bug in the C... | closed | completed | false | 4 | [
"bug"
] | [] | 2022-08-13T06:32:49Z | 2022-12-16T14:57:01Z | 2022-08-13T09:17:04Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | ghost | 10,137 | MDQ6VXNlcjEwMTM3 | User | false |
huggingface/diffusers | 1,496,509,404 | I_kwDOHa8MBc5ZMuvc | 1,703 | https://github.com/huggingface/diffusers/issues/1703 | https://api.github.com/repos/huggingface/diffusers/issues/1703 | Using local dataset but blocked at load_dataset | Hi, @pcuenca @patil-suraj @anton-l
I'm trying to fully fine-tune SD u-net on my own dataset, including about 1M image-text pairs.
I'm following the script in `examples/text_to_image`.
However, it's blocked at load_dataset with very slow processing. It has been 12 hrs but the preparation of these 1M data is still ... | closed | completed | false | 4 | [] | [] | 2022-12-14T12:14:54Z | 2022-12-14T17:02:33Z | 2022-12-14T17:02:33Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | BrandonHanx | 37,724,292 | MDQ6VXNlcjM3NzI0Mjky | User | false |
huggingface/diffusers | 1,496,607,919 | I_kwDOHa8MBc5ZNGyv | 1,704 | https://github.com/huggingface/diffusers/issues/1704 | https://api.github.com/repos/huggingface/diffusers/issues/1704 | question about attention block | why we use torch.baddbmm to do query @ key?
https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention.py#L640 | closed | completed | false | 1 | [] | [] | 2022-12-14T13:14:04Z | 2022-12-28T03:41:20Z | 2022-12-28T03:41:20Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | xiaohu2015 | 16,861,194 | MDQ6VXNlcjE2ODYxMTk0 | User | false |
huggingface/diffusers | 1,496,649,847 | I_kwDOHa8MBc5ZNRB3 | 1,705 | https://github.com/huggingface/diffusers/issues/1705 | https://api.github.com/repos/huggingface/diffusers/issues/1705 | Using services on centos (on server boot) training not working | I have added `train.py` to services, which I've called the `train_dreambooth.py` using `os.system` in `train.py` file, and im getting this error:
```
(diffusers) [ec2-user@ip-172-31-0-133 sd]$ systemctl status trainsd.service -l
● trainsd.service - Train stable diffusion
Loaded: loaded (/usr/lib/systemd/system... | closed | completed | false | 1 | [] | [] | 2022-12-14T13:38:09Z | 2022-12-14T14:22:54Z | 2022-12-14T14:21:45Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | ariand9 | 6,691,197 | MDQ6VXNlcjY2OTExOTc= | User | false |
huggingface/diffusers | 1,497,533,087 | I_kwDOHa8MBc5ZQoqf | 1,708 | https://github.com/huggingface/diffusers/issues/1708 | https://api.github.com/repos/huggingface/diffusers/issues/1708 | How to pass yaml config file for inference for dreambooth and SD2.1 ? | Here is my training configuration :
```
accelerate launch train_dreambooth.py \
--pretrained_model_name_or_path="stabilityai/stable-diffusion-2-1" \
--pretrained_vae_name_or_path="stabilityai/sd-vae-ft-mse" \
--output_dir="./zwx/" \
--revision="fp16" \
--with_prior_preservation --prior_loss_weight=1.0 ... | closed | completed | false | 2 | [
"stale"
] | [] | 2022-12-14T23:00:14Z | 2023-01-23T15:03:37Z | 2023-01-23T15:03:37Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | sifrah | 6,054,235 | MDQ6VXNlcjYwNTQyMzU= | User | false |
huggingface/diffusers | 1,497,728,549 | I_kwDOHa8MBc5ZRYYl | 1,709 | https://github.com/huggingface/diffusers/issues/1709 | https://api.github.com/repos/huggingface/diffusers/issues/1709 | Getting an error with pipeline_onnx_stable_diffusion_img2img.py AttributeError: 'Tensor' object has no attribute 'astype'. Did you mean: 'dtype'? | ### Describe the bug
I'm getting an error when trying to use onnx img2img pipeline
Changing line 353 from image = image.astype(latents_dtype) to image = np.array(image).astype(latents_dtype) fixes this for me and seems to work but not sure if this is correct?
### Reproduction
from PIL import Image
from diffusers... | closed | completed | false | 3 | [
"bug"
] | [] | 2022-12-15T03:33:55Z | 2022-12-16T21:21:13Z | 2022-12-16T14:24:18Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | Justanetizen | 84,889,212 | MDQ6VXNlcjg0ODg5MjEy | User | false |
huggingface/diffusers | 1,497,897,617 | I_kwDOHa8MBc5ZSBqR | 1,712 | https://github.com/huggingface/diffusers/issues/1712 | https://api.github.com/repos/huggingface/diffusers/issues/1712 | For some modules using GPU/CPU setup issues | **What API design would you like to have changed or added to the library? Why?**
API:
UNet2DConditionModel
Reason:
When I use this module for a simple initialization, it tries to use the GPU, which uses about 699m of memory, for example:
unet = UNet2DConditionModel.from_config(unet_config).
I just need an... | closed | completed | false | 3 | [
"stale"
] | [] | 2022-12-15T06:58:49Z | 2023-02-01T15:03:17Z | 2023-02-01T15:03:17Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | Night-Quiet | 73,934,131 | MDQ6VXNlcjczOTM0MTMx | User | false |
huggingface/diffusers | 1,498,317,979 | I_kwDOHa8MBc5ZToSb | 1,715 | https://github.com/huggingface/diffusers/issues/1715 | https://api.github.com/repos/huggingface/diffusers/issues/1715 | LoRA for Dreambooth | Hi,
Have come across LoRA for diffusion models.
It might be right for diffusers to collaborate and later get to one single diffusers SDK as Dreambooth is supported here as well in the current diffusers' library.
Interested to learn about your thoughts on this. 😃
Link: https://github.com/cloneofsimo/lora | closed | completed | false | 13 | [
"stale"
] | [] | 2022-12-15T12:10:29Z | 2023-01-27T17:07:36Z | 2023-01-27T17:07:36Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | polavishnu4444 | 2,263,240 | MDQ6VXNlcjIyNjMyNDA= | User | false |
huggingface/diffusers | 1,498,445,693 | I_kwDOHa8MBc5ZUHd9 | 1,716 | https://github.com/huggingface/diffusers/issues/1716 | https://api.github.com/repos/huggingface/diffusers/issues/1716 | Remove unethically trained diffusion models from model hub | @patrickvonplaten @patil-suraj @anton-l
I wonder what your thoughts are on the arguments the artists are making on Artstation, Twitter and Instagram right now. The main point is that the artists didn't give permission for their work to be scraped and used to create an AI that will threaten their livelihood, without c... | closed | completed | false | 19 | [
"stale"
] | [] | 2022-12-15T13:33:47Z | 2023-01-23T15:03:32Z | 2023-01-23T15:03:32Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | lukovnikov | 1,732,910 | MDQ6VXNlcjE3MzI5MTA= | User | false |
huggingface/diffusers | 1,498,859,125 | I_kwDOHa8MBc5ZVsZ1 | 1,717 | https://github.com/huggingface/diffusers/issues/1717 | https://api.github.com/repos/huggingface/diffusers/issues/1717 | cached models do not load when HF_HUB_OFFLINE=1 | ### Describe the bug
The model hub seems to be having a bad time at the moment. That's a bummer, but I expect my application should still work fine in offline mode. But it didn't automatically fall back to offline mode, and when I manually set the `HF_HUB_OFFLINE` environment variable and tried again, it still didn't ... | closed | completed | false | 4 | [
"bug",
"stale"
] | [] | 2022-12-15T17:56:19Z | 2023-01-16T07:53:21Z | 2023-01-16T07:53:21Z | CONTRIBUTOR | null | 20260407T133413Z | 2026-04-07T13:34:13Z | keturn | 83,819 | MDQ6VXNlcjgzODE5 | User | false |
huggingface/diffusers | 1,337,900,056 | I_kwDOHa8MBc5PvrwY | 172 | https://github.com/huggingface/diffusers/issues/172 | https://api.github.com/repos/huggingface/diffusers/issues/172 | Add options to disable/hide progress bar | **Is your feature request related to a problem? Please describe.**
This is a minor thing, but I find the progress bar annoying when I run inference with pipeline successively.
See this screenshot for example.
<img width="754" src="https://user-images.githubusercontent.com/25161192/184474554-a20cd389-9d11-4a64-bb... | closed | completed | false | 7 | [
"good first issue"
] | [] | 2022-08-13T08:16:23Z | 2022-08-30T10:30:07Z | 2022-08-30T10:30:07Z | MEMBER | null | 20260407T133413Z | 2026-04-07T13:34:13Z | hysts | 25,161,192 | MDQ6VXNlcjI1MTYxMTky | User | false |
huggingface/diffusers | 1,499,427,144 | I_kwDOHa8MBc5ZX3FI | 1,720 | https://github.com/huggingface/diffusers/issues/1720 | https://api.github.com/repos/huggingface/diffusers/issues/1720 | Several Errors when Finetuning Stable Diffusion on M1 Pro | Hello everyone!
I recently tried following [this guide](https://github.com/huggingface/diffusers/tree/main/examples/dreambooth) to finetune stable diffusion.
I get 2 errors.
**First error**
```
AttributeError: Can't pickle local object 'main.<locals>.<lambda>'
```
**Second error**
```
CalledProcessError: Comma... | closed | completed | false | 3 | [
"stale"
] | [] | 2022-12-16T01:42:21Z | 2023-01-23T15:03:30Z | 2023-01-23T15:03:30Z | NONE | null | 20260407T133413Z | 2026-04-07T13:34:13Z | fakerybakery | 76,186,054 | MDQ6VXNlcjc2MTg2MDU0 | User | false |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.