Cond Zero

condzero's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in black-forest-labs/FLUX.2-dev about 10 hours ago

Finetune loss never decrease

1

#36 opened 3 months ago by

fahadh4ilyas

New activity in lldacing/flash-attention-windows-wheel 8 months ago

flash_attn-2.7.4.post1%2Bcu128torch2.8.0cxx11abiTRUE-cp312-cp312-win_amd64

3

#8 opened 10 months ago by

mythicvision

New activity in LifuWang/DistillT5 11 months ago

Is this something you can do for the WAN UMT5EncoderModel?

#3 opened 11 months ago by

condzero

New activity in HiDream-ai/HiDream-I1-Full 12 months ago

Why did the authors choose such a low token max_sequence_count for this model? 128 doesn't cut it. S/B at least 512 tokens. What gives?

#12 opened 12 months ago by

condzero

New activity in stabilityai/stable-diffusion-3.5-large over 1 year ago

Settings?

1

#18 opened over 1 year ago by

brennanisgreat

GPU memory

1

#19 opened over 1 year ago by

gebaltso

New activity in black-forest-labs/FLUX.1-dev over 1 year ago

Unable to run this model. Loading always fails on a hugging face module: modeling_utils.py

5

#49 opened over 1 year ago by

condzero

My version to use Dev and Schnell on a 3090 using quants with a gradio front end.

1

#23 opened over 1 year ago by

NuclearGeek

New activity in An-619/FastSAM over 2 years ago

Some suggestions for the demo!

🤝 1

2

#1 opened almost 3 years ago by

abidlabs