Skip to content
Date and source

Dates follow the news timeline in Beijing time and include both endpoints.

Feb 3Tue
  1. Hugging Face Blog22

    PRX Text-to-Image Model Training Design: Lessons from Ablation Studies

    The Hugging Face blog publishes the second installment of the PRX text-to-image model training series, using ablation studies to document which training techniques actually work. The baseline is PRX-1.2B with 1.2B parameters, using pure Flow Matching, Flux VAE latent space, 256×256 resolution, a global batch size of 256, and 100k training steps, with data consisting of 1 million synthetic images generated by MidJourneyV6.

Jan 27Tue
  1. Hugging Face Blog22

    Architecture Choices in China's Open-Source AI Ecosystem: How MoE, Multimodality, and Domestic Hardware Can Surpass DeepSeek

    Chinese open-source models have nearly unanimously shifted to MoE architecture, with Kimi K2, MiniMax M2, and Qwen3 all adopting this approach, aiming for sustainable deployment at controlled costs. Open-source activity expanded to multimodality and agents starting February 2025, with StepFun's Step-Audio-R1.1 voice model outperforming closed-source models, and Tencent open-sourcing Hunyuan Video and Hunyuan 3D.

Dec 24Wed
Dec 16Tue
Nov 25Tue
  1. Hugging Face Blog60

    Diffusers Integrates FLUX.2 with 4-bit Quantized Loading Support

    Hugging Face Diffusers announces support for FLUX.2, providing invocation code for Flux2Pipeline and Flux2Transformer2DModel. The example uses the diffusers/FLUX.2-dev-bnb-4bit repository with the Mistral3 text encoder and enables enable_model_cpu_offload.

    Why it matters: Diffusers integrates FLUX.2 and provides a 4-bit quantized loading example, enabling creators to evaluate the barrier to local deployment.

Nov 20Thu
  1. Google DeepMind62

    Google Launches AI Image Verification Feature in the Gemini App

    Google launches an image verification feature in the Gemini app. Users can upload an image and ask a question to determine via SynthID digital watermarking whether the image was generated or edited by Google AI. Since its launch in 2023, SynthID has watermarked over 20 billion pieces of AI-generated content. Google plans to extend verification to formats such as video and audio and integrate it with more entry points including Search.

    Why it matters: Google brings SynthID watermark detection into the Gemini app, allowing creators to determine whether an image was generated or edited by Google AI.

  2. Google DeepMind70

    Google DeepMind Releases Nano Banana Pro (Gemini 3 Pro Image) Image Model

    Google DeepMind releases Nano Banana Pro (Gemini 3 Pro Image), a high-fidelity image generation and editing model built on Gemini 3 Pro, now available to developers as a paid preview on Google AI Studio and Vertex AI.

    Why it matters: Gemini 3 Pro Image is available to developers as a paid preview, supporting 2K/4K output and multi-image composition, enabling evaluation of existing image generation workflows.

  3. Google DeepMind82

    Google DeepMind Releases Nano Banana Pro (Gemini 3 Pro Image) Image Generation and Editing Model

    Google DeepMind releases Nano Banana Pro (Gemini 3 Pro Image), built on Gemini 3 Pro, featuring more accurate in-image text rendering, multilingual generation, and real-time information visualization.

    Why it matters: The Gemini 3 Pro-based image model introduces new capabilities in text rendering and multi-image consistency, enabling evaluation of image generation workflows.

Oct 21Tue
  1. Hugging Face Blog40

    Hugging Face AI Sheets Adds Vision Capabilities: Process Images in Spreadsheets with Open-Source Models

    Hugging Face AI Sheets releases a major update adding vision support, allowing users to directly view, analyze, and extract information from images, as well as generate and edit images in spreadsheets without writing code. Powered by Inference Providers, the tool can call thousands of open-source models, supports extracting structured data from receipts, documents, and charts, and can generate images from prompts and produce batch variants. Results can be exported as CSV or Parquet.

Aug 19Tue
Jul 31Thu
  1. Hugging Face Blog30

    Implementing an MCP Server in Python: A Gradio-Based AI Shopping Assistant

    A Hugging Face blog post demonstrates how to build an AI shopping assistant in Python using Gradio's MCP integration, automatically converting Python functions into MCP tools callable by an LLM. The example calls the IDM-VTON virtual try-on model via the Gradio API, combined with VS Code AI Chat and the playwright MCP server to browse web pages, enabling clothing search and try-on effect generation.

Jul 23Wed
  1. Hugging Face Blog62

    Hugging Face Releases a Fast Inference Optimization for Flux LoRA

    Hugging Face releases a LoRA inference optimization for Flux.1-Dev, combining Flash Attention 3, torch.compile, and FP8 quantization to achieve a 2.23× speedup on H100.

    Why it matters: Provides a complete optimization recipe for Flux LoRA inference, including real-world data for hot-swapping and consumer-grade GPUs, which can be directly applied to your own image generation pipeline.

Jun 19Thu
  1. Hugging Face Blog62

    Hugging Face Tutorial: Fine-Tuning FLUX.1-dev with QLoRA on Consumer Hardware

    Hugging Face publishes a tutorial demonstrating how to fine-tune FLUX.1-dev on a single GPU using QLoRA and diffusers, with peak VRAM around 9GB. On an RTX 4090 at 512x768 resolution with LoRA rank 4, training for 700 steps takes approximately 41 minutes, whereas standard BF16 LoRA requires 26GB of VRAM.

    Why it matters: Hugging Face officially provides the complete workflow and measured VRAM data for fine-tuning FLUX.1-dev on single-GPU consumer hardware, which can be adapted for your own style training.

May 21Wed
May 8Thu
Apr 23Wed
  1. OpenAI News62

    OpenAI launches its latest image generation model, gpt-image-1, in the API

    OpenAI announces that its latest image generation model is now available in the API as gpt-image-1, enabling developers and businesses to directly integrate professional-grade, customizable visual content into their own tools and platforms.

    Why it matters: OpenAI has integrated its latest image generation model into the API as gpt-image-1, allowing developers to evaluate image generation integration for their own tools.

Apr 11Fri
  1. Hugging Face Blog22

    Visual Salamandra Released: Extending 7B Multimodal Understanding to Images and Video

    The Language Technology Lab releases Visual Salamandra, extending the capabilities of the Salamandra large model to images and video. Based on a 7-billion-parameter foundation model, it integrates the Google SigLIP-So400m encoder with a 2-layer MLP projector, employing a four-stage training process and a late-fusion architecture.

Mar 25Tue
  1. OpenAI News62

    OpenAI Releases GPT-4o System Card Addendum: 4o Image Generation

    OpenAI releases a GPT-4o system card addendum introducing 4o image generation. This capability is stronger than the previous DALL·E 3 series models, capable of generating photorealistic images and accepting image inputs for transformation.

    Why it matters: OpenAI outlines the capability changes of 4o image generation compared to DALL·E 3 in the system card addendum, providing insight into its positioning for image input and photorealistic output.

Feb 24Mon
Jan 31Fri
  1. Hugging Face Blog22

    AI Art Tools Monthly Report Issue 1: A Review of Key Advances in Image and Video Generation in 2024

    The Hugging Face blog launches the first issue of the AI Art Tools Monthly Report, reviewing key releases in the creative AI space in 2024. In image generation, diffusion models have shifted from U-Net architectures to DiT and flow matching. Flux.1 surpasses Midjourney v6.0 and DALL·E 3 (HD) on multiple benchmarks, while SDXL remains the mainstream foundation for personalized and controllable generation.

Dec 9Mon
  1. Hugging Face Blog22

    Hugging Face community releases open preference dataset for text-to-image

    Hugging Face's Data is Better Together community releases an Apache 2.0 licensed text-to-image preference dataset covering multiple image generation categories, mixing different model families and prompts of varying complexity. The dataset is based on images generated by Flux and Stable Diffusion models, with ready-to-use preference datasets and flux-dev-lora-finetune provided. Preprocessing and post-processing code is open-sourced on GitHub.

Oct 22Tue
  1. Hugging Face Blog62

    Diffusers supports Stable Diffusion 3.5 Large

    Stability AI releases Stable Diffusion 3.5, including an 8B Large model and an 8B timestep-distilled model. The latter supports 4-8 step few-step inference. The models are available on Hugging Face Hub and can be used with Diffusers.

    Why it matters: SD3.5 Large and the timestep-distilled version are both available on Diffusers, with quantized inference and LoRA training paths provided.

Jul 30Tue
  1. Hugging Face Blog38

    Memory-efficient Diffusion Transformer with Quanto and Diffusers

    Hugging Face demonstrates how to use the Quanto quantization tool in the Diffusers library to reduce memory usage of Transformer diffusion models. Using PixArt-Sigma, Stable Diffusion 3, and Aura Flow as examples, quantizing the diffusion backbone and text encoder to FP8 reduces SD3 inference VRAM from 16.403 GB to 8.204 GB, with virtually no quality loss and a slight increase in latency.

Jun 20Thu
Jun 12Wed
  1. Hugging Face Blog62

    Diffusers integrates Stable Diffusion 3 Medium, 2B parameter model available on Hugging Face Hub

    Stability AI's Stable Diffusion 3 Medium (2B parameters) is now available on Hugging Face Hub and integrated with Diffusers, along with Dreambooth and LoRA training scripts.

    Why it matters: SD3 Medium is now available on Diffusers, with VRAM optimization and LoRA fine-tuning scripts included, allowing users to evaluate local image generation costs.

Jun 6Thu
May 16Thu
  1. OpenAI News18

    How Canva built its AI-powered Magic Studio

    Canva is a visual communication platform with over 175 million monthly active users, used for creating presentations, videos, documents, websites, and social media graphics. Facing the challenge that most knowledge workers globally lack design training, Canva enables anyone to create visually appealing content through its user-friendly interface, extensive asset library, and time-saving tools.

Feb 26Mon
  1. Hugging Face Blog22

    Hugging Face Details AI Watermarking Tools and Techniques: From Nightshade and Fawkes to Truepic

    Hugging Face publishes an article surveying watermarking approaches for AI-generated content, distinguishing between two pathways: embedding at generation time and adding after generation. It also introduces watermarking and detection tools available on the Hub. The article mentions image "invisible modification" tools such as Glaze and Photoguard, as well as data poisoning tools like Nightshade and Fawkes, used to prevent models from learning or generating faces. Truepic embeds metadata according to the C2PA standard and combines it with watermarks to enhance provenance verifiability.

Feb 3Sat
  1. Hugging Face Blog62

    SegMoE Released: Building Mixture-of-Experts Diffusion Models from Scratch

    Segmind releases the SegMoE framework, which can create Mixture-of-Experts (MoE) diffusion models from scratch and has been integrated into Hugging Face diffusers. This release includes three model versions—SegMoE-4x2, SegMoE-2x1, and SegMoE-SD4x2—along with the segmoe code package for building custom MoE models. The models are licensed under Apache 2.0.

    Why it matters: The release provides code for building MoE diffusion models from scratch along with three merged versions, enabling creators to combine multiple pretrained models into new ones.

Jan 15Mon
Jan 4Thu
Jan 2Tue
  1. Hugging Face Blog62

    Hugging Face Releases Advanced Training Script for SDXL Dreambooth LoRA

    Hugging Face releases an advanced training script for SDXL Dreambooth LoRA, integrating Pivotal Tuning and the Prodigy optimizer into diffusers, along with reusable training parameters.

    Why it matters: Hugging Face officially integrates Pivotal Tuning and the Prodigy optimizer into diffusers training scripts, providing directly reusable parameters and experimental comparisons.

Dec 5Tue
Nov 9Thu
  1. Hugging Face Blog78

    Using Latent Consistency LoRA to Generate SDXL Images in 4 Steps

    Hugging Face releases the LCM LoRA method, which trains a small number of LoRA adapter layers instead of distilling the entire model, reducing image generation for SDXL and Stable Diffusion from 25 to 50 steps down to 4 to 8 steps.

    Why it matters: LCM LoRA enables SDXL to generate images in 4 steps. The article provides directly runnable diffusers code and speed comparisons across multiple hardware configurations.

Oct 24Tue
Oct 19Thu
  1. OpenAI News62

    DALL·E 3 Launches on ChatGPT Plus and Enterprise

    OpenAI announced that DALL·E 3 is now available in ChatGPT Plus and Enterprise. To prepare for a broader release, OpenAI developed a set of safety mitigations and also shared its progress on content provenance research.

    Why it matters: DALL·E 3 is open to ChatGPT Plus and Enterprise, and the official announcement also explains the safety mitigations made for the expanded release.

Oct 3Tue
  1. Hugging Face Blog32

    Hugging Face Diffusers supports accelerated Stable Diffusion XL inference with JAX on Cloud TPU v5e

    Hugging Face Diffusers now supports running SDXL inference on Cloud TPU v5e via JAX, achieving high-performance, low-cost deployment with JIT compilation and XLA parallelism. The demo generates 4 images of 1024×1024 in parallel on multiple TPU v5e-4 instances in about 4 seconds, with pure generation time of about 2.3 seconds; TPU v5e costs less than half of TPU v4.

Oct 2Mon
Sep 29Fri