DC-Gen is a post‑training framework that accelerates text‑to‑image diffusion models by using a deeply compressed latent space. It first aligns the base model’s latent representations with a lightweight embedding alignment, then applies minimal LoRA fine‑tuning to preserve generation quality. Experiments on SANA and FLUX.1‑Krea show that DC‑Gen‑FLUX cuts 4K image generation latency by 53× on an NVIDIA H100 and, with NVFP4 SVDQuant, achieves a 138× total speedup on a single NVIDIA 5090 GPU.
arXiv:2607. 19064v1 Announce Type: cross Abstract: Large-scale visual generators are increasingly capable but costly to train, fine-tune, and deploy.
Large-scale visual generators are increasingly capable but costly to train, fine-tune, and deploy. We introduce Mage-Flow, a compact 4B-scale generative stack for efficient text-to-image generation and instruction-based image editing.
arXiv:2609.39222v1 Announce Type: new Abstract: High-compression tokenizers are essential for scaling latent image generative models. However, aggressive compression creates a fundamental tradeoff be...
SpeeDiff is a latent diffusion model that jointly trains a VAE and diffusion network from scratch, using a Tweedie Pixel Reconstruction loss to prevent latent collapse. The approach allows scaling a transformer‑based architecture and improves representation alignment. Experiments show the SpeeDiff‑XL model trains 140× faster than Vanilla SiT and 61× faster than REPA, achieving an FID of 1.50 on ImageNet 256×256 and 1.53 on 512×512 without guidance.
arXiv:2610.09450v1 Announce Type: new Abstract: Pixel-space diffusion models avoid the lossy VAE of latent models, which suggests an advantage on downstream tasks where fine-grained detail matters. W...
arXiv:2610.10524v1 Announce Type: new Abstract: Highly compressed video autoencoders offer an effective way to accelerate video diffusion models, as the Diffusion Transformer (DiT) operates on far fe...
The paper revisits pixel-space diffusion models, showing that a vanilla diffusion transformer can generate directly in pixel space by swapping velocity prediction for clean image prediction. It introduces MFlow, an adaptive parameterization that interpolates between these two prediction modes based on data representation and model capacity, and demonstrates competitive performance on ImageNet, UCF‑101, and text‑to‑image tasks. Additionally, the authors add ConvRefiner, a lightweight convolutional module trained post‑hoc to restore high‑frequency details while preserving FID.
Intern Lumina U2 is a diffusion large language model that uses multi-codebook visual tokens in a fully discrete space, allowing each visual position to be represented by a group of complementary codes. The model concatenates embeddings at each position and employs a lightweight autoregressive head to predict codes sequentially while the diffusion backbone processes all positions in parallel. Built on the unified 4D representation of AToken, it supports image, video, and 3D inputs, achieves leading performance in visual understanding, image generation, and editing, and includes a cross‑platform training, inference, and evaluation system with a stability‑aware sampling strategy that speeds up image generation by 4.45×.
arXiv:2606.21030v2 Announce Type: replace-cross Abstract: Diffusion-based image compression has achieved strong perceptual quality at ultra-low bitrates. However, existing codecs are often tied to sp...
Unified multimodal generation covers discrete, continuous, and hybrid paradigms. Discrete approaches with single-codebook visual tokenizers align naturally with text modeling but have limited represen...
ToDoPro introduces an early token‑reduction strategy for reference‑based image generation that uses a few denoising steps on 2× downsampled reference images to probe token importance with the existing Diffusion Transformer. Tokens are then hierarchically pruned: low‑importance tokens are discarded, mid‑importance tokens are replaced by their downsampled versions, and high‑importance tokens are retained. Experiments on three base models and two benchmarks show significant inference speedups, reduced peak GPU memory, and improved generation quality by filtering out noisy reference tokens.
WeLM-Image is a unified image foundation model that combines multimodal understanding with text‑to‑image generation, instruction‑guided editing, and multi‑reference synthesis. It uses a Mixture‑of‑Transformers backbone with Mixture‑of‑Experts modules, enabling 31.7B stored parameters and about 5.4B active generation parameters per latent token. The model, optimized for single‑GPU deployment, achieves top rankings on PRISM‑Bench, Qwen‑Image‑Bench, and UnicBench, and its turbo variant can generate a 1024×1024 image in under five seconds on a single NVIDIA H20 GPU.