Stars
[NeurIPS 2025 Spotlight] LeMiCa: Lexicographic Minimax Path Caching for Efficient Diffusion-Based Video Generation
Taming large-scale full-parameter few-step training with self-adversarial flows! 👏🏻
Official SeedVR2 Video Upscaler for ComfyUI
Official inference repo for FLUX.2 models
Repo for SeedVR2 & SeedVR (CVPR2025 Highlight)
FIBO is a SOTA, first open-source, JSON-native text-to-image model built for controllable, predictable, and legally safe image generation.
Qwen-Image-Lightning: Speed up Qwen-Image model with distillation
[ICCV 2025] Enhancing spatial understanding in text-to-Image diffusion models
An inference and training framework for multiple image input in Flux Kontext dev
Qwen-Image is a powerful image generation foundation model capable of complex text rendering and precise image editing.
Solve puzzles. Improve your pytorch.
Official implementation of "Normalized Attention Guidance"
A course of learning LLM inference serving on Apple Silicon for systems engineers: build a tiny vLLM + Qwen.
ConceptAttention: A method for interpreting multi-modal diffusion transformers.
[NeurIPS 2025] Image editing is worth a single LoRA! 0.1% training data for fantastic image editing! Surpasses GPT-4o in ID persistence~ MoE ckpt released! Only 4GB VRAM is enough to run!
Lumina-mGPT 2.0: Stand-Alone AutoRegressive Image Modeling
An implementation of the CSM(Conversation Speech Model) for Apple Silicon using MLX.
[CVPR 2025] Diffusion Self-Distillation for Zero-Shot Customized Image Generation
TangoFlux: Super Fast and Faithful Text to Audio Generation with Flow Matching
Training-free Regional Prompting for Diffusion Transformers 🔥
SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformer
Official implementation of the paper: "FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models"
🚀 Cross attention map tools for huggingface/diffusers
[🚀ICML 2025] "Taming Rectified Flow for Inversion and Editing" Using FLUX and HunyuanVideo for image and video editing!