repositories Search Results · topic:inference-acceleration org:thu-ml fork:true
Filter by
4 results
inthu-ml (press backspace or delete to remove)
[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-t…
TurboDiffusion: 100–200× Acceleration for Video Diffusion Models
[ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.
SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse–Linear Attention
Sponsor open source projects you depend on
Contributors are working behind the scenes to make open source better for everyone—give them the help and recognition they deserve.Explore sponsorable projectsProTip! Press the / key to activate the search input again and adjust your query.
Sponsor open source projects you depend on
Contributors are working behind the scenes to make open source better for everyone—give them the help and recognition they deserve.Explore sponsorable projectsProTip! Press the / key to activate the search input again and adjust your query.