chengzeyi Comfy-WaveSpeed issues

chengzeyi / Comfy-WaveSpeed

[WIP] The all in one inference optimization solution for ComfyUI, universal, flexible, and fast.

MIT License

641 stars 19 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Bypass on Compile model+ doesn't work

#96 AiD-teng opened 8 hours ago
1
Still the same issue

#95 squallyang520 opened 2 days ago
2
Not work with gguf and instantx ipadapter

#94 jesenzhang opened 3 days ago
1
SDXL. Speedup on default workflow, no speedup if you use lora.

#93 Ph0rk0z opened 4 days ago
4
[Request] Chaining sequence of Apply First Block Cache

#92 Miczu opened 5 days ago
6
"Apply First Block Cache" node does not work with Kijai HunyuanVideo nodes

#91 morganavr opened 5 days ago
3
SDXL model caching not working with Perturbed Attention Guidance

#90 ikarsokolov opened 5 days ago
3
You can suppress this exception and fall back to eager by setting: import torch._dynamo torch._dynamo.config.suppress_errors = True

#89 dernuiJK opened 5 days ago
4
.

#88 shaun-ba closed 5 days ago
0
The settings `start = 0.2`, `end = 0.8`, `max_consecutive_cache_hits = 5` provide better results.

#87 easygoing0114 opened 6 days ago
7
would you keep in mind CPU users please , to endure that can be run using CPU

#86 youforgetsomething closed 3 days ago
5
Dynamic gives an error

#85 GavChap closed 4 days ago
0
Question: FBC support for Stable Diffusion 1.x/2.x works based on SDXL support?

#84 simonlui closed 4 days ago
5
Handle multiple model calls on the same timestep

#83 blepping closed 1 day ago
9
Adding support for Cosmos by Nvidia

#82 ShmuelRonen opened 1 week ago
0
Clean up some of the FBCache exception handling and state tracking

#81 blepping closed 1 week ago
2
plz add scaled soft weights

#80 xueqing0622 opened 1 week ago
1
use flux pulid is wrong

#79 ddddddeyrueuad opened 1 week ago
3
EnhancedCompileModel for MPS

#78 aimerib closed 1 week ago
0
SDXL:No transformer blocks found

#77 kileybeam opened 1 week ago
1
RTX 2060 Support?

#76 krigeta opened 1 week ago
3
DoubleStreamBlock.forward() got an unexpected keyword argument 'attn_mask'

#75 1272870698 closed 1 week ago
2
Error running SamplerCustomAdvanced

#74 hellosky23 opened 1 week ago
6
sdxl Invalid acceleration

#73 huotian233 opened 1 week ago
2
The effect of SD 3.5 is very poor.

#72 zhilengjun opened 1 week ago
4
SamplerCustomAdvanced ：Pointer argument (at 0) cannot be accessed from Triton (cpu tensor?)

#71 wqgalex closed 1 week ago
4
The issue of Wavespeed and Pulid working together

#70 tuolaku closed 1 week ago
7
KeyError: 'text_encoders'

#69 EmmaThompson123 closed 1 week ago
3
Does this node need triton installed?

#68 IAFFeng closed 1 week ago
3
error when i use LTX with compile node, but complie node work with flux

#67 thefirstLeonliao opened 1 week ago
3
I get "velocator is not installed" when trying to use "🚀Load & Quantize Diffusion Model" with Hunyuan Video

#66 ManOrMonster opened 1 week ago
1
Actually, 3090 support fp8e5m2 torch.compile

#65 codexq123 closed 1 week ago
2
RuntimeError

#64 windson-maker opened 1 week ago
3
About SD 3.5 WaveSpeed option

#63 ShmuelRonen opened 1 week ago
2
The node is having no effect in the example workflow for SDXL

#62 danganbenpa opened 1 week ago
12
K采样出错

#61 chenfeng6a closed 1 week ago
3
problem

#60 ZB678232321 closed 1 week ago
2
运行K采样器时出错

#59 JasonKEN432580 opened 1 week ago
9
.

#58 shaun-ba closed 1 week ago
1
batch size and fbc threshold

#57 fgdfgfthgr-fox closed 1 week ago
6
Something wrong with versioning + ComfyManager

#56 shaun-ba closed 1 week ago
1
SDXL demo workflow has an error "No transformer blocks found"

#55 seemon2 closed 1 week ago
3
SDXL compile does not work

#54 fgdfgfthgr-fox closed 1 week ago
1
Error running after update

#53 alex13by opened 1 week ago
1
change prompts cause recompile

#52 wwwffbf closed 1 week ago
2
sdxl , No double blocks found

#51 xueqing0622 closed 1 week ago
2
Before it becomes hot and popular, would you optimize vram for it?

#50 ptmaster opened 1 week ago
24
SDXL Attribute Error 'double_blocks'

#49 ialhabbal closed 1 week ago
6
Fix broken internal link to 'Enhanced torch.compile' section

#48 shinshin86 closed 1 week ago
1
Compile+ takes so long at the beginning to setup that it negates any improvement

#47 shaun-ba closed 1 week ago
2