Welcome back, my AI-obsessed friends.

Shake off the weekend—it’s Monday. This is the latest issue of /Imagine, your go-to source for the freshest news and developments in AI. Here’s EVERYTHING you need to know. Let’s dive in…

🍏 Apple Gets in the LLM Game!

The Bytes:

✣ Apple dropped a paper on their new MM1 family of multimodal models. The largest of which is 30 billion parameters.

✣ According to the paper, The MM1 models excel at tasks like image captioning, visual question answering, and natural language inference, thanks to their diverse training. The benchmarks they shared illustrate that it is comparable to Gemini and slightly behind GPT4. They even acknowledge they used training data generated by GPT4.

✣ Apple’s investment in AI, projected to reach $1 billion annually, underscores its commitment to advancing AI. It’s expected this will get integrated into Siri and the entire Apple app suite.

Why you should care: This will blow regular people’s minds and expose a whole new demographic to AI and its potential. The thing that I’m most excited about is that it will turn Siri into a more powerful and intuitive assistant. This could lead to a new era of AI interaction, where Siri can understand and respond to complex multimodal inputs, making it an even more indispensable tool for Apple users. All eyes are on WWDC in June. I think Apple will surprise us.

🧠 Grok-1 Goes Open Source

The Bytes:

✣ xAI just dropped a bombshell with the open release of Grok-1, a 314 billion parameter Mixture-of-Experts model, under the Apache 2.0 license.

✣ Grok-1, trained from scratch by xAI, is designed for general purposes and not fine-tuned for specific applications yet. Perplexity and others have already shared that they intend to tune it and release it into their products soon.

✣ It’s officially the largest-ever open-sourced LLM.

✣ Developers can access Grok-1’s base model weights and network architecture via GitHub.

Why you should care: Elon Musk's xAI has made a significant move by open-sourcing Grok-1. This milestone offers unprecedented access to a state-of-the-art language model, enabling developers to build upon it and accelerate AI research and application development. I can’t wait to see what people create with it.

🎥 Runway Drops Lip Sync

The Bytes:

✣ RunwayML just rolled out Lip Sync, a feature that animates faces to match any voice lines in uploaded audio clips, creating realistic generative videos.

✣ Lip Sync supports various audio inputs, including existing clips, voice recordings, and new text-to-speech generations.

✣ The tool is optimized to work best with forward-facing, photorealistic faces.

Why you should care: It seems everyone is dropping Lip Sync, but I have to say Runway’s might be the best so far. RunwayML’s Lip Sync is a game-changer for video producers, marketers, and creators. It opens up new possibilities for creative AI storytelling. Check it out, and let me know what you think.

Deep Fake by Argil AI ⚡️

Crazy AI Voice Progress 👀

Google Research → VLOGGER 🪄

New Magnific Style Transfer Tool 🖼️

AI Splat Editing 🛠️

Krea Patch Enhance 🧠

New AI Tool Updates ✨

Prompt share: film still, an astronaut close up, --ar 2:3 --stylize 750

Sharing this one to illustrate, sometimes the simplest of prompts can yield incredible results. Enjoy, my friends. Love you all. Have a great day!✌🏼

P.S. → Letz is now available to all European countries! Give it a try!

Join Us: Don't forget to follow me on Twitter/X at @jake_joseph and the brand on its fresh new page @slashimagineai. Let’s be friends.

Share The Insights: Know someone else who's obsessed with AI? Forward this newsletter and let's expand the community.

Want to advertise in /Imagine’s Daily Newsletter or on our Tools Page? → Book here!

Want more resources? →  Check out our Tools Page!

Feedback

Your perspectives help shape the content and hopefully help me make the newsletter suck less every day. Share your feedback on this issue or what you'd like to see next by directly replying to this email or by reacting to the poll above.