AI changes video angles • FLUX upgrades • AI storyboards • 3D model textures • AI news this week
Welcome to the AI Search newsletter. Here are the top highlights in AI this week.
Tencent has unveiled Hunyuan-Large, an open-source mixture of experts AI model. With only 389 billion parameters and 52 billion activation parameters, it matches the performance of the much larger LLama3.1-405B. Read more
Don’t like reading? Here’s a video overview covering all the highlights in AI this week:
Black Forest Labs has introduced two new modes in FLUX1.1 [pro]: Ultra and Raw Modes. The Ultra mode allows for image generation at 4x the standard resolution, while the Raw mode significantly boosts diversity in human subjects and produces more realistic nature photography. Read more
ConsiStory is a new AI model that can generate consistent images of a subject from different text prompts. This model can generate multiple images of a subject in different settings without requiring any training or optimization. Read more
A good photo on your Linkedin or business profile makes a huge difference. You could do a physical photoshoot, which costs you over $200 and hours posing awkwardly at a camera. Or, with AI Portrait, just upload one photo, and get a portfolio of 50 professional photos in minutes. Save time and money - try it today!
InstantIR is a new AI model that can restore damaged images with incredible detail and realism. It uses a novel diffusion-based method that dynamically adjusts the generation condition during inference, allowing it to produce high-quality restored images. This model can even restore extreme degradation and offer creative restoration options with textual descriptions. Read more
MoGe is a powerful AI model that can turn 2D single images into 3D point maps. This model demonstrates strong generalizability and high accuracy in various tasks, including monocular estimation of 3D point maps, depth maps, and camera field of view. Read more
X-Portrait 2 is a powerful AI model that can animate static portraits with realistic expressions. It uses a state-of-the-art expression encoder model to capture subtle facial expressions from a driving video and generate fluid and expressive videos. Read more
xAlerts is a powerful tool that helps you track the activity of your favorite accounts. Stay informed from the latest activities of investors, celebrities, influencers, athletes, and more. Try it for free!
MVPaint is a new AI tool that can generate high-quality 3D textures for 3D models. It uses a combination of multi-view generation, 3D inpainting, and UV refinement to create seamless and consistent textures. This technology has the potential to revolutionize the field of 3D asset production. Read more
Google's AI model, Big Sleep, has discovered a previously unknown security flaw in SQLite. The model uses advanced variant-analysis techniques to identify similar, potentially exploitable flaws in related code sections. This discovery demonstrates the effectiveness of AI in proactive defense, potentially leading to a significant advantage for defenders. Read more
ACE is an AI model that can generate and edit images based on instructions. It uses a special type of AI called diffusion models and can perform a wide range of tasks, from generating new images to editing existing ones. This makes it a powerful tool for creative tasks. Read more
OmniParser is a new AI method that helps computers understand and interact with user interfaces. It can parse screenshots into structured elements, identifying interactable icons and understanding their semantics. This enables computers to generate actions that can be accurately grounded in the corresponding regions of the interface. Read more
Google has accidentally leaked its AI agent, Jarvis AI, which can browse the web for you in Chrome. Jarvis AI is an AI assistant that can automate web browsing tasks, allowing users to focus on more important things while the AI agent shops or books flights for them. It's expected to be officially unveiled in December. Read more