Midjourney Tests 'Style Consistency' Feature Supporting V6 Version (With Usage Instructions)

Jaaz Open Source AI Design Agent Soars! Batch Image Generation at One Click, Creativity Production Takes Off!

Midjourney Video is Coming! V8 Model is Ready to Launch, A New Era of AI Creativity is Around the Corner!

According to the latest information, Midjourney's video function is about to be launched this month, while the development of V7.1 and V8 models is also accelerating. This article will provide you with a detailed analysis of Midjourney's latest developments, including breakthroughs in video functions, server upgrades, style references (sref) optimization, as well as future model plans, giving you an overview of the cutting-edge progress in AI image generation. The video function is ready to launch, and the rating party will start next week.

NUS Launches OmniConsistency: Achieving Cost-Effective Image Style Consistency to Challenge GPT-4!

OpenAI Makes a Big Move! Image Generation API Upgrade: Real-Time Streaming Preview + Multi-Round Editing MCP Enables Boundless Creation!

OpenAI announced significant updates to its image generation API (Responses API), adding new features such as streaming transmission, multi-round editing, and integration with Model Context Protocol (MCP) tools and real-time web data search. These new features not only improve the efficiency of image generation but also provide developers with more flexible creation methods, marking a new stage for AI-driven visual content creation. AIbase has compiled the core highlights of this update and its far-reaching impact on the industry.

Pixeltorte's Fangtang large model successfully registered as the first image large model to be filed in the domestic imaging industry

Recently, Pixeltorte's independently developed "Fangtang large model" has officially passed the filing by the Cyberspace Administration of China (CAC), becoming the first application-level image large model with official qualifications in the domestic imaging industry. This achievement not only marks a breakthrough in technology for the Fangtang large model but also demonstrates its compliance and safety standards meeting national requirements. The successful registration of Fangtang is an important milestone in the field of artificial intelligence, especially against the backdrop of the increasingly developing image generation technology. Pixeltorte's independent research and development in this technology highlights its strength in AI.

ByteDance Releases Open Source Multi-modal Model BAGE From Image Generation to World Modeling

ByteDance recently officially released its latest open source multi-modal foundation model - BAGEL (Big Advanced Generalized Embodied Learner), starting a new stage for multi-modal AI models with a scale of 7 billion effective parameters. BAGEL performs excellently in key tasks such as image understanding, generation, and editing, and has surpassed current mainstream open source vision-language models (VLM) like Qwen2.5-VL and InternVL-2.5 in multiple standard evaluations.

ByteDance releases 14B parameter multi-modal powerhouse BAGEL, outperforms Qwen2.5-VL in image generation媲美SD3

ByteDance Opensources Multimodal Model BAGEL: New Breakthroughs in Text and Image Generation and Editing

Google Imagen 4 and Imagen 4 Fast Debuts on GCP Vertex Quota Menu: AI Image Generation Reaches New Heights

Drawing and Image Generation! Tencent H混元 Launches Game Visual Generation Platform

Tencent officially released the H混元 Game Visual Generation Platform, which is an AI content engine built on the H混元 large model, specifically designed for industrial-level content production in the gaming industry. The launch of this platform marks the entry of the game art design industry into a new era of efficient creation, expected to increase creation efficiency by tens of times. In the past, game art designers often had to jump between multiple software when creating character images, from finding reference images to drafting sketches, making three-view drawings, and creating dynamic demonstrations, the entire process was scattered and cumbersome, with files requiring repeated imports and exports.