Instructions to use Bl4ckSpaces/Project-Cosmos-V1.3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Bl4ckSpaces/Project-Cosmos-V1.3 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Bl4ckSpaces/Project-Cosmos-V1.3", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
Project-Cosmos-V1.3 (High Fidelity)
Cosmos V1.3 is the direct evolution of the successful V1.1. It corrects the anatomical stiffness of V1.2 by reverting to V1.1's exact structural weights, while significantly improving texture detail.
The Recipe
- Anatomy (Input/Middle): Reverted strictly to V1.1 Settings (Drop 70-80%, Strength 0.5-0.6). This guarantees the return of perfect fingers and poses.
- Style (Output): Drop Rate Reduced to 25% (from 40%). By allowing more fine-grained texture data from NoobAI to pass through (while slightly adjusting strength to 0.75), we achieve a "High Fidelity" look without compromising the structural integrity.
Recommended Settings
- Sampler: Euler a / DPM++ 2M SDE
- CFG: 5.0 - 7.0
- Steps: 28+
- Downloads last month
- 1