1 Repo
Awesome GitHub RepositoriesVRAM Quantization
Reducing precision of specific model components like linear layers to lower video memory usage.
Distinct from Precision Quantization: Specifically targets the reduction of VRAM by quantizing linear layers in transformer blocks, distinct from general bit-width reduction of the entire weight set.
Explore 1 awesome GitHub repository matching artificial intelligence & ml · VRAM Quantization. Refine with filters or upvote what's useful.
Awesome VRAM Quantization GitHub Repositories
Finde die besten Repos mit KI.Wir suchen mit KI nach den am besten passenden Repositories.
- nvlabs/sana
NVlabs/Sana
8,310Auf GitHub ansehenSana is a framework for high-resolution image and video synthesis based on a linear diffusion transformer. It provides a toolkit for the training, fine-tuning, and execution of text-to-image and text-to-video models, as well as a video generative world model capable of simulating physical environments with precise spatial control. The project is distinguished by its use of linear complexity layers to handle high resolutions and its support for long-form, minute-length video generation in real time. It implements a two-stage inference paradigm that separates structural generation from visual t
Lowers memory usage by applying fp8 or fp4 precision to linear layers within transformer blocks.
Python
