SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2312.09608 · NeurIPS · 2024

Faster Diffusion: Rethinking the Role of the Encoder for Diffusion Model Inference

Joost Van De Weijer, Tao Liu, Ming-Ming Cheng, Shiqi Yang, Yaxing Wang, Jian Yang, Fahad Khan, Senmao Li, Taihang Hu, Linxuan Li

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 3 functions out of this paper's own repositories and ran 3 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
hutaiHang/Faster-Diffusion canonical 2 of 2
hutaihang/faster-diffusion canonical 1 of 1
FunctionStatusWhere it lives
get_canny Ran hutaihang/faster-diffusion/utils_if.py
code served (permissive licence) · get_code("23d7f44738089752")
get_promptls Ran hutaiHang/Faster-Diffusion/utils_sd.py
code served (permissive licence) · get_code("284f982c02433d5c")
load_512 Ran hutaiHang/Faster-Diffusion/utils_sd.py
code served (permissive licence) · get_code("c2a212c016115360")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

https://sen-mao.github.io/FasterDiffusion Custom Diffusion 2.42s 41%↓ Custom Diffusion w/ Ours 1.42s ControlNet 3.20s ControlNet w/ Ours 1.52s Stable Diffusion DDIM 2.42s DPM-Solver++ 1.13s DPM-Solver++ 0.64s 41%↓ DDIM w/ Ours 1.42s 43%↓ 24%↓ 20%↓ DeepFloyd-IF DPM-Solver++ 16.09s DDPM 34.55s DPM-Solver++ w/ Ours 12.97s DDPM w/ Ours 26.27s 34%↓ 51%↓ DiT 26.25s DiT w/ Ours 17.35s 32%↓ VideoFusion 1.12s VideoFusion w/ Ours 0.76s * Equal contribution. Author ordering determined by coin flip over a Google Hangout. † The corresponding author. 38th Conference on Neural Information Processing Systems (NeurIPS 2024).

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2312.09608")
get_code_for_paper("2312.09608")
have("2312.09608")

Connect an agent — have() is free.