AUTOMATIC1111 stable-diffusion-webui-tensorrt issues

AUTOMATIC1111 / stable-diffusion-webui-tensorrt

MIT License

310 stars 20 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Advanced Tab not displaying

#87 damionismyname opened 1 month ago
0
Lora Exports are 1kb and don't work

#86 SamoraHunter opened 3 months ago
0
tensorrt errors ,tab not appearing

#85 Rinox89 opened 4 months ago
0
Anyone get TensorRT Extension running with usefull image sizes ?

#84 XPHUPZHJ opened 4 months ago
3
RuntimeError

#83 KelvinSoh8 opened 6 months ago
1
going higher than 768x768 gives an error but if batch size is 2 then one image gens correctly

#82 Dargunsh opened 6 months ago
0
ERROR:root:Failed to build engine: [Errno 2] No such file or directory

#81 Dargunsh closed 6 months ago
1
Going above 2048x2048 with tensorrt using hires_fix possible?

#80 PmNz8 opened 7 months ago
0
new version

#79 kalle07 opened 7 months ago
0
RuntimeError: cpu gpu

#78 kalle07 opened 7 months ago
0
Input shape must be divisible by 64 in both dimensions

#77 NoteToSelfFindGoodNickname opened 8 months ago
1
Trying to create an rttensor for a lora, I got the error ERROR:root:Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument mat1 in method wrapper_CUDA_addmm) Traceback (most recent call last):

#76 left1000 opened 8 months ago
6
I seem to somehow end up stuck in a situation where I cannot generate images at all, it claims no profile can be found, despite me using the same settings that worked moments ago.

#75 left1000 opened 8 months ago
4
Making a profile can freeze forever.

#74 left1000 opened 8 months ago
0
How do I delete a tensorrt profile? If I delete the file it continues to show up in the webgui, and throws errors, because it realizes a profile is missing.

#73 left1000 opened 8 months ago
5
Generate Default Engines

#72 Cigoler closed 8 months ago
2
How do I create an engine profile for adetailer? or is rttensor incompatiable with adetailer?

#71 left1000 opened 8 months ago
3
I can't create a custom TensorRT Engine Profile with a custom name, I can merely open advanced settings and alter one of the names that already exists.

#70 left1000 closed 8 months ago
0
from what I can tell install.py requires importlib_metadata but pip is installing importlib_metadata by the name importlib-metadata and so it can't be found? also is my cpu being treated as a secondary cuda choice and that is ruining the code? is there anyway to tell this extension to ignore my cpu's built in graphics functionalities that I never use?

#69 left1000 opened 8 months ago
8
Cannot choose between profiles

#68 sfingali opened 8 months ago
2
Official TensorRT extension from NVIDIA is available now

#67 Jonseed opened 8 months ago
1
Need the feature to load LoRA and base model seperately

#66 FlamingJay opened 9 months ago
0
can i run the official webui in code?not the UI?

#65 henbucuoshanghai opened 9 months ago
0
inpainting models don't work

#64 yuvraj108c closed 7 months ago
0
Can't use Hires Fix

#63 xameryn opened 10 months ago
0
amd gpu

#62 303Aki303 opened 10 months ago
1
urgent: SDXL TensorRT for guinea pig photo production

#61 rustrust closed 10 months ago
0
(today) https://huggingface.co/stabilityai/stable-diffusion-xl-1.0-tensorrt

#60 LockMan007 opened 10 months ago
0
Cannot run img2img on SDXL

#59 aferlo78 opened 10 months ago
0
SDXL Support

#58 CyberTimon opened 11 months ago
3
not working on SDXL

#56 Shaistrong opened 11 months ago
0
Cast binding shape to tensorrt.Dims before setting

#55 rajeevsrao closed 10 months ago
0
Runtime Error - FAILED TensorRT.trtexec

#54 78Alpha closed 11 months ago
2
Can't use any trt that isn't 512x512 batch 1

#53 BuffMcBigHuge opened 11 months ago
2
Does webui have a plan to support onnx runtime for stable diffusion accerlating?

#52 yjymickey opened 12 months ago
0
Does SD model matter?

#51 dxwarlock opened 12 months ago
1
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument mat1 in method wrapper_CUDA_addmm

#50 AdamBlazer opened 1 year ago
3
how can i increase Maximum prompt token count?

#49 yjymickey closed 9 months ago
0
unable to use ControlNet extensions when tensorRTs are in use

#48 randomname1212 opened 1 year ago
1
stable-diffusion-webui-tensorrt\scripts\trt.py", line 6, in <module> from modules import script_callbacks, sd_unet, devices, shared, paths_internal ImportError: cannot import name 'sd_unet' from 'modules' (unknown location)

#47 lucasjinreal opened 1 year ago
2
TensorRT official webui

#46 lucasjinreal opened 1 year ago
3
How to clear pytorch base model GPU memory? Tensorrt GPU memory larger than pytorch.

#45 WudiJoey opened 1 year ago
1
Added Batch Processing, Shape appending, Dynamic alert for too large of a shape for converting trt models.

#44 jebarpg opened 1 year ago
0
Anyone managed to setup/use on linux?

#43 yuvraj108c closed 1 year ago
9
"System cannot find path specified" when trying to convert to onnx

#42 DuckersMcQuack opened 1 year ago
2
SD Web UI - Loading Old Prompts During Generation.

#41 ClipSkipper opened 1 year ago
0
What is meaning "Loras are baked in into the converted model. "?

#40 edisonchan opened 1 year ago
3
tensorrt-8.6.1-cp310-none-linux_x86_64.whl is not a supported wheel on this platform.

#39 yuvraj108c closed 1 year ago
1
Cannot generate at higher res than 512x512

#38 ellugia closed 1 year ago
3
Uncaught exception detected: Unable to open library: nvinfer_plugin.dll

#37 deeyonn opened 1 year ago
7