TurboDiffusion: ×ばつ Acceleration for Video Diffusion Models
rcm distillation video-generation mlsystem diffusion-models inference-acceleration ai-infra consistency-model sageattention sparse-linear-attention
-
Updated
Aug 27, 2026 - Python
#
sageattention
Here are 13 public repositories matching this topic...
Language: All
Filter by language
Sort: Most stars
Sort options
[ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.
attention vit quantization video-generation mlsys inference-acceleration ai-infra vision-transformer sparse-attention llm sageattention
-
Updated
Feb 25, 2026 - Cuda
Cross-platform installer for Triton and SageAttention on ComfyUI. Simplifies GPU-accelerated inference setup for Windows users with automated dependency management and RTX 5090 support.
windows automation installer cuda pytorch triton gpu-acceleration windows10 build-tools cli-tool windows11 stable-diffusion comfyui rtx-5090 sageattention
-
Updated
Aug 11, 2026 - Python
NVIDIA Sol-Attn for ComfyUI / Triton kernel on SM89 - SM121, with zero-copy MiniMax H3 nodes: memory-efficient attention, scheduled tau with graph preview, and feed-forward chunking. Measured ×ばつ vs SageAttention and −37% MLP peak VRAM on H3
triton attention video-generation memory-optimization diffusion-models blackwell memory-op sparse-attention comfyui sageattention comfyui-custom-nodes trito minimax-h3 sol-attn
-
Updated
Aug 13, 2026 - Python
🪟 为 Windows AI 开发者提供预编译 wheel 文件的集中仓库 | 自动抓取并整理 PyTorch、Flash Attention、xformers、SageAttention 等常用库的最新版本 | 免编译,开箱即用 | 特别适合 ComfyUI 和 Stable Diffusion 用户
-
Updated
Sep 6, 2026
Bleeding-edge ComfyUI for NVIDIA DGX Spark (GB10/Blackwell/sm_121a). CUDA 13 + SageAttention v3 (sm_121a) + NVFP4 + 14 custom-node packs + Flux 2 Dev / LTX 2.3 22B / ACE-Step v1.5 XL Turbo pre-bundled with abliterated text-encoder paths.
-
Updated
Jun 28, 2026 - Shell
Automatically benchmark and optimize attention in diffusion models. 1.5-2x speedup on RTX 4090.
flux performance optimization attention diffusion stable-diffusion comfyui flash-attention comfyui-custom-node sageattention
-
Updated
Feb 9, 2026 - Python
Selective fused FP8 checkpoint + ComfyUI runtime for LingBot-Video Dense 1.3B. Measured 1.27x in a local one-step RTX 5080 sampler smoke test; T2V, TI2V, experimental FLF.
-
Updated
Aug 1, 2026 - Python
An all-in-one docker image that runs the latest ComfyUI with SageAttention.
-
Updated
Sep 6, 2026 - Shell
Automated Windows installer for ComfyUI with NVIDIA GPU optimizations
python windows automation powershell scaffold windows-10 pytorch batch automatisation uv windows-11 comfyui sageattention
-
Updated
Sep 1, 2026 - PowerShell
Modular ComfyUI optimizations for MiniMax H3: fused NVFP4 MLP, low-memory Sage2, CAB low-step sampling, long-sequence VRAM safeguards, benchmarks and workflows.
-
Updated
Aug 12, 2026 - Python
MiniMax H3 SageAttention & Spectrum workflow pack for ComfyUI, providing ready-to-run API JSON workflows for text/image/reference-to-video rendering with clean output naming and low-diff, reproducible setup.
spectrum optimization optimisation video-generation comfyui comfyui-workflow sageattention minimax-h3
-
Updated
Aug 4, 2026
Add this topic to your repo
To associate your repository with the sageattention topic, visit your repo's landing page and select "manage topics."