Official codebase for "Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation"

Advancing Open-source World Models

High Performance LLM Inference Operator Library

Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.

Towards Real-Time Diffusion-Based Streaming Video Super-Resolution — An efficient one-step diffusion framework for streaming VSR with locality-constrained sparse attention and a tiny conditional de…

Combine all open source AI image-generated models and video-generated models, to generate AI videos in predefined workflow easily.

A general fine-tuning kit geared toward image/video/audio diffusion models.

Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.

Nano vLLM

[ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.

PromptEnhancer is a prompt-rewriting tool, refining prompts into clearer, structured versions for better image generation.

TurboDiffusion: 100–200× Acceleration for Video Diffusion Models

Tile-Based Runtime for Ultra-Low-Latency LLM Inference

Enjoy the magic of Diffusion models!

MoBA: Mixture of Block Attention for Long-Context LLMs

Allow torch tensor memory to be released and resumed later

Official implementation of BLIP3o-Series

flex-block-attn: an efficient block sparse attention computation library

StreamDiffusion, Live Stream APP

ComfyUI custom node for lightx2v

Tile primitives for speedy kernels

Code for Draft Attention

Official implementation of paper "VMoBA: Mixture-of-Block Attention for Video Diffusion Models"

LTX-Video Support for ComfyUI

A minimal implementation of DeepMind's Genie world model

📚A curated list of Awesome Diffusion Inference Papers with Codes: Sampling, Cache, Quantization, Parallelism, etc.🎉