- AI Toast
- Posts
- Gemini Omni Prompt Guide
Sponsored by
Welcome to AI Toast!
Here’s what’s on the menu today:
Google Shares Gemini Omni Prompt Guide
OpenAI Cuts GPT-5.6 Prices
Super Tools for Designers
Tutorial: How to Create Music with AI
Quick Toasts: NVIDIA stock climbs as AI infrastructure demand remains strong
Total read time: About 5 minutes, perfect for a quick coffee break
Image Source: google blog
Google has published an official Gemini Omni Prompting Guide, sharing best practices for creating high-quality AI-generated videos with natural language prompts.
More Detail = Better Results: Google recommends describing camera angles, motion, lighting, style, location, and actions for more precise video generation.
Edit with Conversation: Instead of rewriting prompts, you can simply ask Gemini Omni to change backgrounds, objects, camera angles, captions, or animations through natural language.
Smarter World Knowledge: Gemini Omni uses Google's world knowledge to create realistic scenes without requiring highly detailed technical prompts.
Prompt camera movements like dolly zoom, close-up, over-the-shoulder, smartphone style, or one continuous shot.
Mix Multiple Inputs: Combine text, images, videos, and audio in a single prompt to generate richer, more consistent scenes.
Keep Characters Consistent: Use reference images or videos to maintain the same characters, objects, and environments across edits.
Personal Take: Google is making AI video creation much more intuitive. Rather than relying on complex prompt engineering, Gemini Omni encourages creators to refine videos through natural conversation, making professional-quality video generation more accessible. Full guide.
Stop googling AI tools at midnight.
There are over 3,000 of them in The Shift’s vault, already vetted, ready to explore.
Subscribe and get instant access to the tool vault, a 1000+ prompt library, and free AI courses built for people with actual work to do.
Plus, the daily newsletter that keeps you sharp on everything moving in AI in under 5 minutes a day.
They’re also giving away a free 1-year Claude Pro subscription to 3 subscribers.
Subscribe for free to enter. All free. All in one place.
Image Source: OpenAI blog
OpenAI has announced major updates to the GPT-5.6 model family, cutting prices for Luna and Terra while introducing a new Fast mode for GPT-5.6 Sol.
GPT-5.6 Luna, OpenAI's fastest and most affordable model, now costs 80% less.
GPT-5.6 Terra receives a 20% price cut, making everyday AI workloads more cost-effective.
API users can now access GPT-5.6 Sol with up to 2.5× faster response times at 2× the Standard processing price, with no change in intelligence.
Luna and Terra now consume fewer credits, allowing users to get more value from their existing quotas.
Auto-review in ChatGPT and Codex CLI has been upgraded from GPT-5.4 to GPT-5.6 Luna, reducing costs by about 10× for agentic workflows.
OpenAI says ongoing efficiency improvements allow customers to run more AI workloads at lower cost while maintaining high performance.
Personal Take: This isn't just a price drop, it's a strategic move to make advanced AI affordable enough for large-scale production use. Lower costs, faster inference, and more efficient models will make it easier for businesses to deploy AI across everyday workflows.
Superhero Tools (Designers)
Adobe Firefly: An AI creative suite that generates images, vectors, videos, and design assets from text prompts while integrating seamlessly with Adobe Creative Cloud.
Figma AI: Build with AI from end-to-end. Explore. Use the Figma agent to generate new design directions, create diagrams, edit images, search your files.
Magnific: The complete platform of creative AI tools for image, video, and audio generation. Create anything from campaigns, product shots to filmmaking.
Google stitch: Stitch generates UIs for mobile and web applications, making design ideation fast and easy.
Framer AI: Framer Agents turn prompts into native, editable layers on your canvas. Prompt, refine and ship production websites without leaving your canvas.
Tutorial: How to Create AI Music with Lyria 3.5
Google DeepMind has integrated Lyria 3.5 into Google Flow Music, making AI music generation more accurate, expressive, and customizable.
How to use it
Visit Google Flow Music and sign in.
Click New Track.
Describe the music you want (genre, mood, instruments, vocals, BPM).
Generate your track with Lyria 3.5.
Edit your prompt if needed and regenerate.
Export your finished song or download individual stems for further editing.
Why Lyria 3.5?
Better prompt understanding
More expressive vocals
Richer, more natural arrangements
Set exact BPM and export stems for full-length songs
Pro Tip: Include details like genre, BPM, mood, instruments, and vocal style in your prompt for the best results.
Quick AI News Bites
NVIDIA stock climbs as AI infrastructure demand remains strong. Investors responded positively after Microsoft's earnings reinforced expectations for continued AI data center spending, with NVIDIA's next-generation AI chips remaining in high demand.
UNESCO releases a regional AI assessment for creative industries. The report explores how generative AI is transforming cultural and creative sectors while emphasizing the need for ethical AI governance and responsible adoption.
Anthropic reveals findings from internal AI cybersecurity testing. The company disclosed that several Claude models accessed real systems during controlled security evaluations, highlighting the importance of stronger safeguards for frontier AI models.
Boost revenue and gain new customers by partnering with us
Reach over 35K AI enthusiasts with your product.
Join our newsletter to connect with tech professionals, investors, engineers, managers, and business owners worldwide. DM now!
Got feedback, a story worth toasting, or a wild tech question? I’d love to hear from you; just reply, or find me on X.
Cheers,
— Poonam Soni