Hugging Face's logo
Linoy Tsaban's picture
162 257 779

Linoy Tsaban PRO

linoyts
EmilyWitko's profile picture Fataj's profile picture denofiend's profile picture
·

AI & ML interests

None yet

Recent Activity

liked a model about 6 hours ago
Skywork/SkyReels-V2-T2V-14B-720P-Diffusers
reacted to a-r-r-o-w's post with 🔥 1 day ago
Caching is an essential technique used in diffusion inference serving for speeding up image/video generations. Diffusers just added support for another caching method: First Block Cache - a technique developed by @chengzeyi building upon the ideas of TeaCache. The idea in short is: if the model predictions do not vary much over successive inference steps, we can skip certain steps where the prediction difference is small. To figure out whether an inference step will make a significant improvement to the overall velocity/noise prediction, we calculate the relative difference of the output of the first transformer block at timestep $t$ with $t-1$, and compare it against a selected threshold. If the difference is lower than the threshold, we skip the step. A higher threshold will lead to more steps being skipped. However, skipping many steps is bad because it can throw off the model predictions, and so we need to test and select the threshold based on level of quality-speed tradeoff for every model we use it with. Diffusers usage with CogView4: ```python import torch from diffusers import CogView4Pipeline from diffusers.hooks import apply_first_block_cache, FirstBlockCacheConfig pipe = CogView4Pipeline.from_pretrained("THUDM/CogView4-6B", torch_dtype=torch.bfloat16) pipe.to("cuda") apply_first_block_cache(pipe.transformer, FirstBlockCacheConfig(threshold=0.2)) prompt = "A photo of an astronaut riding a horse on mars" image = pipe(prompt, generator=torch.Generator().manual_seed(42)).images[0] image.save("output.png") ``` Below, you'll find the benchmarks and visualizations of the predicted output at different blocks of the Flux DiT. Docs: https://huggingface.co/docs/diffusers/main/en/optimization/cache PR: https://github.com/huggingface/diffusers/pull/11180 References: - First Block Cache: https://github.com/chengzeyi/ParaAttention - TeaCache: https://github.com/ali-vilab/TeaCache
upvoted a collection 1 day ago
LTX-Video models
View all activity

Organizations

Hugging Face's profile picture 🧨Diffusers's profile picture Hugging Face Internal Testing Organization's profile picture Huggingface Projects's profile picture Snap Research's profile picture Weizmann Institute of Science's profile picture Editing Images's profile picture leditsplusplus's profile picture Latent Consistency's profile picture Editing Audio's profile picture Women on Hugging Face's profile picture +RAIN film festival's profile picture diffusers-internal-dev's profile picture rnri-inversion's profile picture Snapchat Inc.'s profile picture Latent Explorers's profile picture open/ acc's profile picture RF Inversion's profile picture FlowEdit's profile picture LTX Collaborations's profile picture CRINGE's profile picture Réflexion IA's profile picture IP Composer's profile picture Transformers Community's profile picture Inference Endpoints Images's profile picture Hugging Face MCP Course's profile picture Kontext Community's profile picture