-
Notifications
You must be signed in to change notification settings - Fork 492
Insights: NVIDIA/Cosmos
Overview
-
0 Active pull requests
-
- 0 Merged pull requests
- 0 Open pull requests
- 8 Closed issues
- 11 New issues
Could not load contribution data
Please try again later
8 Issues closed by 4 people
-
7B Diffusion video2world model - fine tuning error
#144 closed
Mar 11, 2025 -
Bugs in Tokenizer Training
#141 closed
Mar 9, 2025 -
Questions about action-conditioned Cosmos
#134 closed
Mar 7, 2025 -
AttributeError: module 'cv2.dnn' has no attribute 'DictValue'
#131 closed
Mar 7, 2025 -
Custom Prompt Post-Training Error
#128 closed
Mar 7, 2025 -
Action control WFM release
#126 closed
Mar 7, 2025 -
Bug in create_condition_latent_from_input_frames?
#129 closed
Mar 6, 2025 -
missing pynvml in requirements.txt
#121 closed
Mar 6, 2025
11 Issues opened by 11 people
-
running Post trained video2world model
#145 opened
Mar 11, 2025 -
I want to convert my final checkpoint into a .pt file.
#143 opened
Mar 11, 2025 -
Conditioning with a start image and end image?
#142 opened
Mar 10, 2025 -
failed to construct or call DiTDataModule Video2world post-training
#140 opened
Mar 9, 2025 -
Post-training with one A100-80GB GPU for Cosmos-1.0-Diffusion-7B-Video2World
#139 opened
Mar 8, 2025 -
Multi-Device post training
#138 opened
Mar 7, 2025 -
Data preparation in Tokenizer training
#137 opened
Mar 7, 2025 -
Executing the demo generated a video output with full-black frames
#136 opened
Mar 7, 2025 -
What is the cleanest way to introduce new weights when post-training?
#135 opened
Mar 7, 2025 -
Adversarial Loss for Cosmos Tokenizer
#133 opened
Mar 6, 2025
8 Unresolved conversations
Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.
-
Possibly incorrect augmented latent noise during Diffusion finetuning
#118 commented on
Mar 6, 2025 • 0 new comments -
Dimension mismatch during inference
#122 commented on
Mar 6, 2025 • 0 new comments -
Hanging during model instantiation
#124 commented on
Mar 6, 2025 • 0 new comments -
Error when resuming post-training
#127 commented on
Mar 7, 2025 • 0 new comments -
DGX NGC container error: cannot import name 'AttnBackend' from 'megatron.core.transformer.enums
#132 commented on
Mar 9, 2025 • 0 new comments -
Loss spikes during video2world diffusion model post-training
#130 commented on
Mar 10, 2025 • 0 new comments -
Inference with fine-tuned Video2World-Diffusion Error
#105 commented on
Mar 10, 2025 • 0 new comments