DiffuseKronA: A Personalized Diffusion Model Fine-Tuning Method that Reduces Parameters and Synthesizes High-Quality Images

Major Breakthrough! MiniMax Launches New Image Generation Model, Image-01, at 1/10th the Cost

ByteDance Releases Infinity: A New Breakthrough in Auto-Regressive Text-to-Image Generation, Outperforming Diffusion Models

In the field of artificial intelligence, ByteDance's commercial technology team presents the latest achievement, the Infinity model, which has become a new leader in the auto-regressive text-to-image generation domain with its outstanding performance and innovative technology. This newly open-sourced model not only surpasses Stable Diffusion 3 in image generation quality but also demonstrates significant advantages in inference speed. The core innovation of the Infinity model lies in its use of a Bitwise Token auto-regressive framework, which predicts the next level of resolution by +1 or -1.

Text-to-Image Generation is Here! Zhichu's Open Source Image Generation Model CogView3 is 10 Times Faster than SDXL!

Recently, Zhichu AI has open-sourced its latest creation—CogView3 and its upgraded version CogView-3Plus-3B, injecting new vitality into the field of text-to-image generation. The debut of CogView3 is undoubtedly a significant milestone. As the first model to achieve relay diffusion in the text-to-image generation domain, it employs a unique cascade diffusion method. This innovative approach first generates low-resolution images and then completes the final output through relay-based super-resolution technology. This not only greatly enhances...

AI Daily: The New Rising Star in Text-to-Image Generation? The Mysterious Blueberry Model Emerges; Alibaba Mama Open Sources AI Image Restoration Model; Google Spends $2.7 Billion to Rehire AI Experts

Welcome to the 【AI Daily】 section! Here is your daily guide to exploring the world of artificial intelligence. Every day we present you with the hottest topics in the AI field, focusing on developers, helping you gain insights into technological trends and understand innovative AI product applications. Check out the latest AI products: https://top.aibase.com/ 1. The mysterious Blueberry model emerges: Is it the new leader in AI text-to-image generation or just a marketing gimmick? Considering restructuring into a profitable company to enhance transparency and employee mobility.

Visual Effects Expert Takes Over as CEO of Stability AI to Drive New Developments in Text-to-Image Generation

UK-based AI startup Stability AI has appointed Prem Akkaraju as its new CEO. Akkaraju previously served as the CEO of visual effects company Weta Digital and is also one of Stability AI's investors. The investor lineup for Stability AI is impressive, including former Facebook president Sean Parker among others, with total investments exceeding 1.

Hugging and Intel Release Consistency Solution for Text-to-Image Generation - SPRIGHT

The solution enhances the spatial consistency of text-to-image generation models, significantly improving the models' understanding of spatial relationships in prompts. It develops datasets and methodologies that reach industry-leading levels, creating a large-scale dataset, SPRIGHT, focused on spatial relationships to enhance the performance of text-to-image generation models through experiments and testing.

Microsoft Open Sources Innovative LoRA Combination Method to Enhance Control Over Complex Details in Text-to-Image Generation

LoRA is widely used in the text-to-image generation field, capable of rendering and merging specific elements in images, such as characters and clothing. Researchers have developed the Multi-LoRA combination method, allowing the integration of multiple LoRAs without fine-tuning, while maintaining weight integrity. LoRA Switch excels in image synthesis quality, while LoRA Composite performs exceptionally well in image quality generation. LoRA Switch ensures the rendering of each element by switching between different LoRA models.

SD3-Turbo: A Fast Text-to-Image Generation Model Based on LADD

SD3-Turbo is a text-to-image generation model based on LADD technology. The SD3-Turbo model generates high-quality images through four steps of unsupervised sampling. The LADD technology simplifies the training process and improves performance. The release of SD3-Turbo will bring new breakthroughs to the field of image generation. The SD3-Turbo model improves production efficiency and holds significant importance.

AI Real-time Painting System StreamMultiDiffusion Supports Local Brush and Prompt Generated Images

The AI real-time painting system StreamMultiDiffusion supports local brushing and prompt-generated images. It introduces a multi-prompt stream batch processing architecture to achieve faster panoramic image generation speeds. Key technologies include Latent Pre-Averaging, Mask-Centering Bootstrapping, among others. A new concept, Semantic Palette, is proposed, allowing users to generate high-quality images in real-time.

Stable Diffusion 3 Model Release: Architecture Details Revealed, Is It Helpful for Reproducing Sora?

The Stable Diffusion 3 model introduces the DiT (Diffusion Transformer) architecture, which significantly improves quality. Stable Diffusion 3 outperforms other text-to-image generation systems. The parameter count of Stable Diffusion 3 ranges from 800M to 8B. The SD3 architecture is developed in collaboration with core members of Sora and an assistant professor from New York University.