⚡️ Stability AI's Breakthrough

Good Morning. Imagine creating sophisticated 3D models from a simple text description and a single image. Stability AI makes this a reality with the debut of its innovative Stable Zero123 model—ushering in a new era for creators and designers. Let’s dive in.

Today’s Highlights:

• OpenAI Startup Fund launches Converge 2

• Google introduces AI music composer, MusicFX

• Australian scientists to unveil brain-mimicking supercomputer

DEEP DIVE

Stability AI Unveils Breakthrough in Text-to-3D Imagery

Imagine transforming a single image into a full 3D model with just a text description. That's the impressive new leap in image-to-3D technology announced by Stability AI. Their latest innovation, Stable Zero123, revolutionizes the process by generating multiple images of an object from different viewpoints, which can lead to a detailed 3D model.

To outdo its predecessor, Stable Zero123 taps into an expansive and quality-assured 3D dataset, Objaverse. Training this new model incorporates images and camera angles that greatly bolster the model's predictive capabilities.

What's the secret sauce? Training efficiency. Stability AI reports a 40-times improvement compared to the older Zero123-XL, thanks to better data and optimization in training with larger batches.

While this model opens new vistas for researchers, Stability AI clarifies that Stable Zero123 is currently for research use only—not commercial exploitation. Aspiring commercial users need to take it up with Stability AI directly.

For creators eager to explore this new model, Stable Zero123 comes with a guide on HuggingFace, coupled with the threestudio framework. While the VRAM demands match those needed for Stable Diffusion 1.5 during the new view generation, expect heavier VRAM usage—around 24GB for crafting 3D objects.

Will Stable Zero123 redefine how we create virtual worlds and digital assets? It’s not usable for commercial projects just yet, but it's certainly fueling the imagination for what's on the horizon in 3D model generation.

Keep reading here.

PUNCHLINES

Chatbot CPA: H&R Block launches AI tax filing assistant that will answer users’ filling questions.

Sistine Code: Vatican says AI is the ‘highest-stakes gamble of our future.’

Jamming with Jargon: Spotify confirms test of prompt-based AI playlists feature.

Backdrop Bonanza: Instagram introduces GAI-powered background editing tool.

Banking on Borders: Kapital raises $165M to expand banking platform.

TLDR

Google unveils MedLM for healthcare AI: Google's MedLM, built on the improved Med-PaLM 2, introduces AI models for the healthcare industry with 85% accuracy—on par with medical specialists. Two versions cater to complex and scalable tasks, now available to US Google Cloud customers, with a Gemini upgrade on the horizon.

New fuel-free power technology unveiled at Cop28: South Korea's SEMP debuts AISEG, an AI-based electromagnetic generator, promising eco-friendly and efficient energy without fuel. Promoting constant output and versatility, it aims to power everything from homes to mobile phones, potentially revolutionizing sustainable energy.

Microsoft Expands Azure with Llama 2 and GPT-4 Turbo: Microsoft introduces Meta's Llama 2 and OpenAI's GPT-4 Turbo with Vision to Azure AI Studio, diversifying its AI offerings. Customers can access various AI models, including Llama-2's text and chat options, along with GPT-4's image analysis, as model-as-a-service.

Intel Core Ultra launches for AI-enhanced PCs: Intel unveils Core Ultra processors, designed for AI-enhanced PCs, providing improved power efficiency and advanced graphics. Over 230 AI PC models from leading manufacturers will feature these new chips, promising a transformative AI PC experience and a forecast of dominating 80% of the market by 2028.

TRENDING TOOLS

🧠 MindOne 0.6: Enhance your user interface experience with sophisticated AI technology

🌐 Buenoverse: Construct and navigate shared 2D virtual worlds with ease

💲 PriceGPT: Use AI to scrutinize and optimize your pricing page layout

🎙️ Audio Note AI: Effortlessly convert spoken words into professional text documents

🤝 Final Round: Ace online interviews with real-time transcriptions and AI-driven contextual support

That’s all for today—if you have any questions or something interesting to share, please reply to this email. We’d love to hear from you!

P.S. If you want to sign up for the Supercharged newsletter or share it with a friend, you can find us here.