We’re wrapping up winter with another major update.

What’s new:

1. Seedream 5.0 is live. Now more consistent, smarter, and supports multi-reference for better control over your results.

2. Gemini 3.1 now supports documents. You can upload and work with CSV, PDF, MD, and TXT files directly inside your workflow.

3. VEO now supports 4K. Create higher-quality videos with sharper output and more production-ready results.

4. Four new ready-to-use templates:

• Animated banner with text
• UGC generator (face + voice + product in frame)
• Combine multiple products into one natural composition
• Realistic product rotation animation

🟣 Jump in and try it out 🟣

We’ve rolled out major upgrades across video, motion control, and quality. Here’s what’s live today:

🎬 Runway 4.5

• Create vertical (9:16) and horizontal (16:9) videos
• Generate videos even without an input image
• More flexibility for social, ads, and cinematic formats

🔊 Kling 3

• Now supports sound generation
• Add elements, first & final frames for stronger control
• Up to 15 seconds video length
• Adjustable frame duration in settings

🎥 Kling Omni v3

• Perfect for video generation from references
• Advanced video editing capabilities
• Adjustable frame duration
• Sound support included

🎞 Kling Motion (Motion Control)

• Transfer motion from a source video to your own character
• Up to 30 seconds video support
• Powerful character animation in seconds

🖼 Remove Background Node – Updated

• Higher quality output
• More advanced settings
• Greater control for production-ready results

📺 VEO 4K Upgrade

• 4K quality now available for VEO 3.1 and 3.1 Fast

⚡ Hover Upscale Video

• One-click video upscaling
• Just hover over your video and press Upscale

Hey Sumo-lings

We’re back on AppSumo and wanted to share a quick update. Over the last months, we focused on shipping exactly what you asked for: better models, more practical templates, and clearer onboarding.

New Tier 5

Based on your feedback, we added Tier 5 for power users:

• More credits
• Save up to 40 projects
• Built for larger workflows and experimentation
• 
  

New & upgraded AI models (your top requests)

Since November, we’ve added and improved some of the most requested models:

• Qwen Image – much better quality, editing + LoRA, multilingual prompts (and now cheaper)
• FLUX 2 – realistic object physics, up to 10 references
• Seedream – up to 4K quality, great for high-detail visuals
• Wan Image & Wan Video – high-quality images and video with native audio
• Kling 2.6/Omni – motion control, multi-reference, longer videos
• Nano Banana v3 – multi-reference, manual aspect ratios, up to 4K
• Ideogram – consistent characters & brand visuals
• Gemini 3 Pro (Text) + GPT-5 / 5.1 / o in the Text node

New templates (real workflows, not demos)

We also shipped ready-to-use templates you asked for:

• Product try-on (beauty & lifestyle)
• Product animation for e-commerce
• Product placement in realistic interiors
• Multi-object scene composition
• Talking Characters (image → video + audio)

Better onboarding

You told us it should be easier to start — so we:

• Added more tutorials
• Linked a 2-minute explainer video directly on the AppSumo page

We’ll keep shipping fast and listening closely. Thanks for helping us shape Phygital+ 💜

Major New Feature: VEO 3.1 Now Live!

We’re thrilled to announce that VEO 3.1, Google’s best-in-class video generation model, is now available on Phygital - across ALL account tiers.

Our Commitment to Fair Pricing

We’ve heard your feedback about credits, and we want you to know: we’re listening and we care.

How We Keep Prices Fair

-Real-time price monitoring: We actively track when AI model providers reduce their pricing

Recent examples:

• VEO Fast: Price reduced to reflect lower API costs
• Gemini models: Continuously optimized pricing as Google adjusts rates

To all our AppSumo tier holders: You’re part of the Phygital family, and we’re in this for the long haul.

• All tiers will continue to be supported - no exceptions
• New models will be added regularly - you’ll get access as we expand

Improving Output Quality

We know that sometimes AI models don’t deliver exactly what you’re looking for. We’re tackling this head-on with three major improvements:

1. Enhanced ML algorithms: Smarter model routing and parameter optimization

2. Advanced prompting techniques: Better prompt engineering under the hood

3. AI Design Agents: Intelligent assistants that help you craft better requests and achieve your creative vision

Our goal is to make Phygital not just a tool, but a true creative partner that understands what you want to achieve.

Your voice matters. Every feature improvement, every price adjustment, and every new model we add is influenced by your input. Please continue sharing your thoughts, suggestions, and creative work with us.

Thank you for being part of the Phygital journey!

We’ve just added three powerful new models to Phygital+ — and they’re available on all plans.

Sora 2 (OpenAI) – Generate ultra-realistic video scenes from text or images. Exceptional motion, depth, and cinematic detail — perfect for storytelling, ads, or product showcases.

Runway Aleph – The latest generation of Runway’s video model — faster, smoother, and more consistent. Great for motion design, marketing clips, and creative animations.

Krea – The best model yet for realistic face generation. Ideal for portraits, campaigns, and lifestyle visuals — delivering human-like quality with natural lighting and expressions.

🟣 All available now in your workspace — no extra setup needed. Just open your generator, pick a model, and start creating magic!