The latest AI image models, generators, editing tools, and research for creating and understanding visual content.
Midjourney has introduced a new feature in its V8.1 draft mode that lets users generate 24 images with different styles by adding --sref random to their prompt. Draft mode can be enabled by clicking the ⚡ icon or by including --draft in the prompt.
Why it matters: This update allows users to quickly explore a wide variety of styles within draft mode.
Simon Willison ported the Moebius 0.2B lightweight image inpainting model to run in the browser using WebGPU, removing the need for PyTorch and CUDA. The demo is available at simonw.github.io/moebius-web/. The project was completed as a side task while waiting for an AI coding agent to finish a larger refactor.
Why it matters: This shows that efficient AI models can be made accessible directly in the browser, lowering hardware barriers and broadening access to AI image editing tools.
PaddlePaddle has released PP-OCRv6 on Hugging Face, a multilingual OCR system supporting 50 languages. The model series ranges from 1.5 million to 34.5 million parameters, offering scalable accuracy and efficiency.
Why it matters: This release provides a versatile, open-source OCR solution for 50 languages, enabling developers to choose a model size that fits their deployment constraints.
Midjourney has launched a draft mode for V8.1 that generates 24 low-resolution images per job. Users can select images to render at full quality using the "Vary" option. Draft jobs consume half the fast hours of standard generations.
Why it matters: This update reduces computational cost for initial exploration, making iterative image generation more efficient for users.
Midjourney has updated its default model from V7 to V8.1, following user testing and feedback. The new model offers improvements in coherence, prompt adherence, and text rendering.
Why it matters: This update establishes a new standard for AI image generation quality for all Midjourney users.
Midjourney has improved its conversational mode for both text and voice inputs. Voice sessions now have access to Image Prompts, Style References, sidebar settings, and recent jobs. Image Prompts can be used from the tray and sidebar, and tray images persist across voice submissions until removed.
Why it matters: These updates streamline the creative workflow by integrating voice and image inputs more seamlessly, making it easier for users to iterate on designs without losing context.
Allen AI has released OlmoEarth v1.1, an updated family of Earth observation models with improved efficiency. These models are designed for satellite imagery analysis and are available on Hugging Face.
Why it matters: This release supports advancements in AI for environmental monitoring and geospatial analysis by offering more efficient models.
Google DeepMind is expanding access to Project Genie, a tool that simulates real-world places, to Google AI Ultra subscribers globally. The expansion includes a new capability powered by Street View, enhancing the realism of generated environments.
Why it matters: This integration leverages Google's Street View data to create more accurate and immersive virtual environments for a broader audience.
Midjourney is preparing major aesthetic updates for versions 8.1 and 8.2 and is asking users to help rank images at full 2K resolution for the first time. This user-driven ranking process aims to improve the model's image quality.
Why it matters: User participation in high-resolution image ranking could lead to significant improvements in the visual quality of future Midjourney releases.
Midjourney has released V8.1 Alpha, the latest version of its V8 model, following feedback from V8.0. The update maintains a consistent aesthetic similar to V7 and introduces enhancements to moodboards and srefs functionality.
Why it matters: The update offers users improved creative tools and consistency in AI-generated imagery.
Stability AI has announced Brand Studio, an end-to-end creative production platform powered by a brand's own assets. The platform is designed to streamline content creation while maintaining brand consistency.
Why it matters: This launch signals Stability AI's move beyond model provision into comprehensive creative solutions for brands.
Midjourney has enabled Relax mode for V8 Alpha, making it available to all Standard, Pro, and Mega subscribers. The mode supports all commands except for using --hd and --q 4 at the same time.
Why it matters: This update allows users to generate images in V8 Alpha without consuming fast hours, making the feature more accessible for those with limited GPU time.
Midjourney has released V8.1 on both Discord and midjourney.com, featuring improved sharpness and image quality. The enhancements are especially noticeable for style references (SREFs) and moodboards, but apply to all images.
Why it matters: This update directly improves image quality for all Midjourney users, particularly enhancing style consistency features.
Meta's new Muse Image AI tool can generate photos of users with public Instagram profiles by default, using their faces from posts. Privacy advocates warn users to check their settings, as there is no notification when posts are used in the AI model.
Why it matters: This development raises significant privacy concerns, as users' images may be used in AI-generated content without their knowledge or explicit consent.