Skip to content
Date and source

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

Sep 22Tue
  1. ArchiGen AI · 图像生成与ComfyUI实战30

    How to Mark Locked, Adjustable, and Replaceable Parts in Architectural Renderings Before Your First ComfyUI Enhancement

    Before your first ComfyUI enhancement, attach a one-page protection brief to each architectural rendering, dividing the image into three zones — locked, adjustable, and replaceable — and listing verifiable architectural facts. Locked zones preserve roof edges, window openings, columns, stairs, property boundaries, and camera position. Adjustable zones are limited to materials, exposure, sky, and glass tint. Replaceable zones regenerate vegetation, people, and distant backgrounds according to boundary rules. Run only one operation at a time, and keep a comparison table of the source image and each checkpoint.

Sep 21Mon
  1. note · ryo1144 · 图像视频创作分享27

    100 AI Illustration Generation Prompts [Latest September 2026] — Compatible with ChatGPT, Gemini, Midjourney, and Firefly

    A collection of 100 copy-ready AI illustration generation prompts for ChatGPT, Gemini, Midjourney, and Adobe Firefly, covering scenes such as characters, landscapes, anime, products, blogs, and SNS. The article argues that 2026 image generation places more emphasis on clearly stating purpose, subject, composition, and atmosphere in natural language rather than stacking long incantations like "8K" or "masterpiece," and provides a six-element template along with image correction instructions.

  2. Clip设计夹 · 创作教程与项目分享36

    Meitu Design Studio Uses AI to Batch-Test Styles from a Single Image, Handling the Full Set of Materials for Autumn Collection Launches

    Meitu Design Studio launched a clothing try-on feature that lets users upload a single flat-lay product image to batch-generate try-on renders with different models, poses, and scenes. It also supports AI clothing color change, outfit replacement, and model and scene switching. Users can also generate a full set of images with a single sentence in the homepage dialog box, or upload a competitor's hit image to replicate its lighting and composition 1:1, then continue generating sales videos with scripts automatically produced by an agent team. The tool has been among the first to integrate with mainstream large models such as WorkBuddy and Codex, and supports mobile app operation.

  3. Wireflow · AI短剧与视频制作教程20

    Best White-Label AI Image Software Tools of 2026: A Roundup

    Wireflow ranks first in the 2026 white-label AI image software review, offering 30+ image models via REST API (including Recraft V4, Flux 2 Pro, Nano Banana, Seedream, and Ideogram), with watermark-free output, usage-based billing, and no per-seat fees. Bria AI excels in licensed-data training and IP indemnification, with local deployment support; Scenario can train custom models based on your own asset library.

Sep 20Sun
  1. 王若风 · 开源创作项目分享62

    Qwen Open-Sources Qwen-Image-2.1: 7B Unified Text-to-Image and Image Editing with ComfyUI Support

    The Qwen team open-sourced Qwen-Image-2.1, unifying text-to-image and image editing into a single 7B parameter checkpoint, with native support for 2K generation and RGBA transparent output with alpha channel, up to 10 reference images, and local editing via circle, doodle, or independent mask selection.

  2. Oimi · 中文图像视频教程62

    AI Figure Prompts: 10 Copy-Ready Collector-Grade Designs (2026)

    Oimi published 10 copy-ready AI figure prompts that turn selfies into collector-grade toy designs with acrylic display cases and blister packaging, covering material routes including clay, vinyl, resin, plush, and boxed formats. The author provides quick instructions: copy the prompt and attach a clear photo of yourself; if the prompt doesn't mention reference images, add a face-lock phrase at the beginning to preserve facial features.

  3. 通义千问 · 图像与音频模型62

    Qwen Open-Sources Qwen-Image-2.1 and Image Editing Prompt Rewriting Model

    Qwen open-sourced Qwen-Image-2.1, a unified text-to-image and image editing model with a 7B parameter, 32-layer Single-Stream DiT for visual generation. This version supports generating standard and transparent RGBA images, up to 10 reference images, local editing via circle selection or masks, and maintains consistency for characters and products.

    Why it matters: Qwen open-sourced a unified text-to-image and image editing model, along with a prompt rewriting model, allowing creators to evaluate how to integrate editing workflows.

  4. 通义千问 · 图像与音频模型62

    Tongyi Qianwen Open-Sources Qwen-Image-2.1 Image Generation and Editing Model and Prompt Rewriting Model

    Tongyi Qianwen open-sourced Qwen-Image-2.1, a unified text-to-image and image editing model with a 7B parameter, 32-layer Single-Stream DiT for visual generation, balancing generation quality, inference efficiency, and versatility.

    Why it matters: Tongyi Qianwen open-sourced an image generation and editing model, along with a prompt rewriting model, allowing creators to evaluate image generation and editing workflows.

Sep 19Sat
  1. ArchiGen AI · 图像生成与ComfyUI实战38

    A two-input controlled test in ComfyUI using line art and clay passes from a locked camera

    The tutorial proposes exporting two passes, line art and clay/simple material, from the same locked camera in ComfyUI, and using a ControlNet two-condition controlled test to see which input protects which architectural facts. The method fixes the checkpoint, VAE, sampler, seed and resolution, changing only the conditioning path, and runs four cases: clay alone, line art alone, equal-weight pairing and tuned pairing, with four seeds per case.

  2. Wireflow · AI短剧与视频制作教程22

    Best batch image generation tools of 2026: a comparison of 8 including Wireflow, OpenArt and Recraft

    2026 batch image generation tools fall into three categories: batch prompt apps, template renderers and API platforms. Wireflow, rated best overall, links multiple image models on a node canvas and runs batch jobs through a single REST endpoint; OpenArt supports CSV/TXT batch generation of up to 500 images, the highest free batch limit; Recraft offers an asynchronous batch API at about $0.035 for raster images and about $0.08 for vector images.

Sep 18Fri
  1. Oimi · 中文图像视频教程62

    10 Popular Portrait Poster Prompts (2026)

    This tutorial compiles 10 portrait poster prompts, covering styles such as Japanese ink-wash minimalism, Y2K collage, silhouette narrative, underground club, cinematic red-and-blue color scheme, and typographic outlines; each comes with a complete English prompt and adjustment tips. The author suggests pasting the prompts into ChatGPT or Gemini and replacing the bracketed placeholders; if you need to preserve your own face, you can add an identity instruction at the beginning and attach a clear frontal photo. Use ChatGPT for long, structured text prompts, and Gemini for scenarios with high facial-similarity requirements.

  2. Wireflow · AI短剧与视频制作教程34

    Best AI Fashion Model Generators of 2026: Hands-On Test of 8 Tools Including Wireflow, Modelia, and FASHN AI

    Using the same batch of 5 garments (a printed T-shirt, a pleated skirt, a knit cardigan, a leather jacket, and running shoes), 8 AI fashion model generators were hands-on tested on garment fidelity, model consistency, batch throughput, and price acceptance. Wireflow ranked first for chaining garment preprocessing, model generation, and upscaling into a single canvas pipeline.

  3. 优设 · 创作者专栏34

    How to make AI posters look high-end? 4 pitfalls to avoid to escape a cheap tech feel

    AI posters look cheap mostly because of four basic problems: style words that are too vague, a messy information hierarchy, too many decorative elements, and unclear prompts. The article breaks down these common issues and offers fixes: control the amount of information, establish clear primary and secondary elements, make "high-end feel" concrete, and write poster prompts in order.

Sep 17Thu
  1. ArchiGen AI · 图像生成与ComfyUI实战28

    A reusable ComfyUI workflow should withstand four different types of architectural input

    A ComfyUI workflow that can generate a single architectural rendering may just be tuned to one screenshot; ArchiGen suggests testing portability with four clearly different inputs: a clean clay look, a glass-curtain-wall interior, a low-contrast dusk view, and a cluttered urban environment, which respectively expose problems such as fabricated facade details, depth errors, fake openings, and noise leaking into structural control.

Sep 16Wed
  1. Promptsref.com · 创作分享22

    3 illustration SREF codes to make Midjourney works feel more hand-drawn

    Promptsref.com shares 3 Midjourney illustration SREF codes, respectively featuring sparse, imperfect line art; rough, doodled geometric shapes; and a gothic dark style blending Japanese manga with retro posters. All three build a hand-drawn feel with black lines, warm paper tones, and a small amount of accent color, suitable for children's books, indie comics, posters, stickers, and game key art. Two of them are member-exclusive SREFs.

Sep 15Tue
  1. 张小闲 · 创作者专栏62

    Generate poster title art text with a single prompt, covering styles such as brush calligraphy, jelly letters, and neon.

    The author put together a playbook for generating poster title art text with a single prompt, writing the theme, font feel, color, and atmosphere into the prompt to get a first version out in just over ten seconds. The author tried directions including handwritten brush calligraphy, pink-orange jelly letters, purple flowing light, tech speed letters, futuristic neon, and premium gold lettering, covering scenarios such as e-commerce posters, event key visuals, product detail pages, and AI design covers; the prompt structure is not complicated, and it can be reused by swapping out the theme and style descriptions.

Sep 10Thu
  1. Luma40

    Luma Layers launches text extraction feature

    Embedded text is exactly why you're still opening Photoshop. Layers can now extract it. Select a layer, click Extract, and the copy becomes live text, already automatically matched to the closest font in the font library. Retype. Change the font. Re-layout. Changing a line of copy takes two clicks instead of regenerating. Layers, powered by Uni-1.

  2. Luma58

    Luma Agents launches GPT-Image-2.5, integrates two new OpenAI image models

    Luma announced that GPT-Image-2.5 is now live in Luma Agents, alongside integration of two new OpenAI image models, both available the same day. Among them, Flare targets speed and batch work, while Sunburst targets edits that require precise placement; users can bring in reference images, change only what needs changing while preserving the rest, and then move into video.

  3. Google AI Blog60

    Google DeepMind uses AI to reconstruct a 70-year love story frame by frame

    Google DeepMind partnered with Primordial Soup to release the documentary short "Love, Rendered," using AI to reconstruct the never-filmed meeting scenes of Burt and Ethelle Shatz.

    Why it matters: Google DeepMind used image restoration and performance capture models to reconstruct unfilmed memories, offering a concrete look at how generative tools are used in documentary imagery.

Sep 8Tue
Sep 2Wed
  1. Google AI Blog62

    Google releases Google Pics, an image creation and editing tool

    Google launched Google Pics, an image creation and editing tool under Workspace, built on the Nano Banana image generation and editing model. It will roll out over the coming weeks to all Google AI Pro and Ultra subscribers and most Workspace business customers.

    Why it matters: Google is bringing image generation and editing tools into Workspace apps, giving teams a basis to decide whether to migrate their image production workflows.

Aug 25Tue
  1. Hugging Face Blog62

    Gradio launches gr.Workflow: turning AI pipelines into a drag-and-drop canvas with auto-generated APIs

    Gradio's built-in gr.Workflow describes AI pipelines as graphs of typed nodes, offering a drag-and-drop canvas where each node can run and intermediate results are visible; the same graph is also a REST API and can be deployed to Hugging Face Spaces with one click.

    Why it matters: Gradio turns multi-step AI pipelines into a visual graph and automatically exposes REST interfaces, giving teams a basis to assess migration costs against their existing production workflows.

Jul 23Thu
  1. Hugging Face Blog62

    Diffusers natively supports Nunchaku 4-bit diffusion inference

    A Hugging Face blog post explains that Diffusers now natively supports Nunchaku's 4-bit diffusion inference; loading a quantized checkpoint only requires calling from_pretrained(), with no local CUDA compilation needed.

    Why it matters: Nunchaku's 4-bit quantized inference is now natively integrated into Diffusers, allowing creators to evaluate the trade-off between VRAM usage and speed for image generation on consumer-grade GPUs.

Jul 6Mon
  1. Hugging Face Blog22

    PRX Part 4: Our Data Strategy

    The fourth installment of the PRX series reveals its data pipeline: pre-training data is composed of a mix of public and internal datasets, with VLMs used to regenerate long image descriptions, which are then converted into corpora suitable for streaming training. The team switched to the Qwen3-VL text encoder and computes text latents in real time during training, with measured throughput loss of approximately 3–4% (roughly one extra day for a 30-day training run). Images are uniformly encoded at JPEG quality 92, and initial re-encoding is nearly imperceptible in practice.

Jul 1Wed
  1. Google DeepMind80

    Google DeepMind Releases Nano Banana 2 Lite and Gemini Omni Flash

    Google DeepMind has released two models: Nano Banana 2 Lite (gemini-3.1-flash-lite-image) and Gemini Omni Flash (gemini-omni-flash-preview).

    Why it matters: The two models focus on speed and cost for image generation and conversational editing for video, respectively, allowing developers to evaluate integration into combined workflows.

Apr 21Tue
Apr 10Fri
Feb 27Fri
  1. Google DeepMind74

    Google DeepMind Releases Nano Banana 2 (Gemini 3.1 Flash Image) Image Model

    Google DeepMind releases its latest image generation model, Nano Banana 2 (Gemini 3.1 Flash Image), bringing Nano Banana Pro's advanced world knowledge, quality, and reasoning capabilities to Flash speed.

    Why it matters: Nano Banana 2 brings Pro-level world knowledge and text rendering down to Flash speed, with multi-platform availability and pricing information, enabling assessment of whether to switch image workflows.