Issues
- 0
- 1
error with Mochi loader "vae_stats.json" ?
#95 opened by MushroomFleet - 0
No module named 'triton.common'
#94 opened by abozahran - 0
Is there a way to be able to use mochi fine tune .safetensor inside of comfy ui wrapper
#93 opened by TheMindExpansionNetwork - 20
Missing Node Types
#89 opened by hosseinbehrad - 4
doesn't want to generate video
#92 opened by SheprStone - 1
- 0
Is it possible to use this through the cli?
#77 opened by radna0 - 3
TypeError: object of type 'float' has no len() when running a Mochi node / ComfyUI
#88 opened by netw1z - 4
- 1
Not available in ComfyUI Manager.
#58 opened by Ainaemaet - 8
Try unloading all models before VAE if OOM
#73 opened by Jonseed - 1
- 0
Allocation on device
#87 opened by gioxyer - 0
- 2
MochiVAEEncoderLoader 'layers.0.weight'
#83 opened by sd2530615 - 4
MochiSampler generates only 13 frames on RTX 4070 12GB. How to increase frames amount?
#81 opened by aluap99 - 1
Clip and Motion Module locations and DL link
#84 opened by GoZippy - 6
The output is an incorrect video of pink squares.
#82 opened by mikheys - 2
vae encoder has a problem
#61 opened by mikeyimer - 8
- 1
Documentation: Any Volunteers?!
#79 opened by mmtmn - 2
Triton on Win11 is not detected
#78 opened by aluap99 - 2
- 1
Workflow for IMAGE to VIDEO, Do you have??
#71 opened by ClothingAI - 2
- 0
Deterministic Results?
#76 opened by samhodge-aiml - 0
Hi,bro
#75 opened by T8star1984 - 1
object of type 'float' has no len()
#60 opened by mikeyimer - 8
help! # ComfyUI Error Report ## Error Details Exception Type: RuntimeError ## User specified an unsupported autocast device_type 'mps'
#49 opened by Xyok2022 - 2
manual cast: torch.float32 + unet unexpected: ['scaled_fp8'] ( despite using the right models)
#72 opened by moreAImore - 3
- 2
I get out of memory with more than 61 frames no matter if tiling vae or not
#64 opened by reydeljuego12345 - 2
- 1
Mochi Sigma Schedule adds an extra spurious step?
#69 opened by horsten - 1
Mochi Model Loader doesn't work but down(load) does, because of vae_stats_path
#63 opened by cchance27 - 5
Error in Mochi VAE Decode Spatial Tiling Node on M1 Max - “Input type (c10::BFloat16) and bias type (float) should be the same”
#43 opened by kiminih - 20
MochiSampler using Triton : ImportError: DLL load failed while importing __triton_launcher
#62 opened by Djanghost - 0
'NoneType' object is not subscriptable
#59 opened by NeilWang079 - 17
Why is the effect of using this node for inference different from the original effect of genmoai?
#42 opened by zishen-ucap - 2
- 0
Interleaved instead of tiled VAE decode?
#57 opened by horsten - 2
- 1
Adding "start_at_step" and "end_at_step" options to MochiSampler like in KSampler (Advanced)
#56 opened by horsten - 2
- 13
Workflows are not loading in comfyui
#46 opened by fahdmirza - 18
- 1
VAE Decode Spatial Tiling Node (unknown)
#50 opened by MushroomFleet - 0
- 1
After ~4 clips generated torch.compile stops working with error message: torch._dynamo hit config.cache_size_limit
#44 opened by jepjoo