Post-training with Tinker

Official codebase for "Self Forcing: Bridging Training and Inference in Autoregressive Video Diffusion" (NeurIPS 2025 Spotlight)

Taming large-scale full-parameter few-step training with self-adversarial flows! 👏🏻

Wan2.2-Lightning: Speed up wan2.2 model with distillation

Light Video Generation Inference Framework

The world's first open-source multimodal creative assistant This is a substitute for Canva and Manus that prioritizes privacy and is usable locally.

Official inference repo for FLUX.2 models

HunyuanVideo-1.5: A leading lightweight video generation model

[CVPR-2024] Official Pytorch implementation of "Misalignment-Robust Frequency Distribution Loss for Image Transformation"

[ICCV2025]LeanVAE: An Ultra-Efficient Reconstruction VAE for Video Diffusion Models

pi-Flow: Policy-Based Few-Step Generation via Imitation Distillation

Krea Realtime 14B. An open-source realtime AI video model.

Optimal transport tools implemented with the JAX framework, to solve large scale matching problems of any flavor.

Kandinsky 5.0: A family of diffusion models for Video & Image generation

HunyuanImage-3.0: A Powerful Native Multimodal Model for Image Generation

Official Code for "Rethinking Diffusion Model in High Dimension"

Flash Attention Triton kernel with support for second-order derivatives

code based for rectified flow

HunyuanImage-2.1: An Efficient Diffusion Model for High-Resolution (2K) Text-to-Image Generation​

Official Github Repo for Neurips 2024 Paper Immiscible Diffusion: Accelerating Diffusion Training with Noise Assignment

Foundation Model for Multiplex Spatial Proteomic Images

FACM: Flow-Anchored Consistency Models

TeEFusion: Blending Text Embeddings to Distill Classifier-Free Guidance (ICCV 2025)

PyTorch native post-training library

Industry-level video foundation model for unified Text-to-Video (T2V) and Image-to-Video (I2V) generation.

PyTorch re-implementation for MeanFlow

JAX implementation of MeanFlow