Reading · PicAisso · RSS Amplifier

Hi friends! 👋The AI world moves so fast, with new stuff dropping almost every day. But remember, chase tools that actually boost your workflow, not just the shiniest new toy. Take those fancy pop-out car door handles, they’re cool to look at, but totally useless in real life. Same with some AI features: flashy but forgettable. The good news? Models are getting cheaper and better all the time, like Google’s Veo 3 slashing prices to make pro videos way more accessible. Let’s dive in!⚡

Check out this 80s AI lo-fi I created with Producer AI while you’re reading this newsletter! 🎶😉

I just put together this video using Hailuo AI’s Start and End Frame, and just like that, it whipped up a smooth animation connecting them. Every image in the video? All from Hailuo too. So fast and exactly what I pictured. Then I jumped into Microsoft Clipchamp to piece it all together. After that, I took the video to ElevenLabs and added sound effect and music using their new Studio 3.0. The new Studio 3.0 is next-level. You just type in what you want for SFX, music or audio, and it generates it on the spot. No more digging through sound libraries! I’m sharing a screenshot so you can see how easy the Studio interface is.

Cool AI brands! Advertise in the PicAisso AI art newsletter and on the PicAisso.xyz website! Reach AI creative tools enthusiasts and get your cool AI app noticed. ⚡️

Yes, I am in!

Wan is shipping like 🔥lately. They just dropped Wan 2.2 not long ago, and now here’s Wan 2.5, like Veo3! It can crank out videos up to 10 seconds long, with audio generation that’s perfectly synced to the frames, and it handles both I2V and T2V like a dream. This clip right here was my very first go at text-to-video with it. You can log in to claim free credits daily or generate with relax mode without credits but with a longer waiting time. It’s super generous for testing out all these features.
Prompt: A skateboarder cruising through a vibrant urban park, weaving between blooming flowers and dodging playful dogs under the warm glow of the morning sun. The skateboarder jerks into action, performing a series of ollies and grinds to evade the pursuer, speeding past graffiti-covered walls and over makeshift ramps. As the chase intensifies, he dive through a narrow alley, narrowly avoiding a swinging door, and burst into a bustling marketplace. He weaves between market stalls, skimming past hanging fruits and startled vendors, the clatter of their board echoing off the cobblestone.
I just tried Reve AI’s new edit feature, and it’s way better than Nano Banana for image editing. The revamped UI and workflow are so smooth, perfect for AI tools. I got early access, and wow, the precision blew me away. You can tweak specific parts of an image with simple prompts, and it’s spot-on. I’m sharing a quick screen recording to show it off. Seriously, check out Reve for editing, it’s a game-changer.

Mochi from Genmo AI is back with an updated video model! The video quality is excellent, with natural, realistic movement. The barista’s milk pour is 99% accurate to real latte art, and the foam’s movement, pushing the crema aside, is impressive!

This is such a fun tool, Boba Anime is totally changing how we create anime. With just a simple prompt, you can whip up full anime clips complete with animation, sound effects, and voice over. You can check out more anime short clips here. I tried it out and couldn’t believe how quick it was to get something polished. If you’re into anime or quick video experiments, you have to give it a shot.✨
Prompt: Sailor Moon transforming in a sparkling, moonlit cityscape, yelling, ‘Moon Prism Power, Make Up!’ as her outfit changes in a burst of light. She strikes a pose, holding her tiara, and declares, ‘In the name of the moon, I’ll punish you!’ Dramatic shoujo art style, magical girl transformation sequence, emotional voiceover.

HeyGen launched their AI video agent. I had early access and tested this amazing feature. This video was created with just a prompt using the AI video agent! (no editing at all!) Check out the prompt here. I am impressed with the animation and the visual effect. Just head to HeyGen, click Video Agent in the left sidebar, and type your prompt. The generated video is fully editable.

Decart just launched Lucy-14B, the fastest large I2V model out there, it takes only 6 seconds to generate a video. You can run it on Fal, and it’s really incredible how quick and smooth it is. They also dropped Lucy Edit, an open-source nano banana for videos! 🤯I had early access, and my mind was totally blown, you can edit with text prompts like changing outfits or scenes while keeping the motion perfect. It is Runway Aleph, but cheaper. You get 2000 credits on every sign-up on Decart, and editing a video costs 150 credits. This company’s got so many cool features lined up for launch, and I’m excited to share more as I test their new stuff. 😉

Tencent’s Hunyuan team built SRPO (Semantic Relative Preference Optimization), a fine-tuning method and model built on FLUX.1-dev for super realistic text-to-image generation. It’s excellent at creating lifelike images, nailing details like skin and hair while capturing imperfections perfectly. Try this image model on Fal. You can prompt the AI to add skin imperfections to your AI art prompt for that perfect-yet-imperfect look!😉
I know we’ve got lots of AI agents today, but this one’s a standout for music lovers, Tunee AI. With just a prompt like “late night lo-fi at a Shanghai street,” I got this full video in less than 10 minutes. It’s super collaborative, chatting with you for input along the way and offering options to make decisions easier if you’re stuck on choices. You can pick the music style, video style, and aspect ratio, plus it even writes the storyboard and camera movements for you. I love how it speeds up the whole creative process without feeling overwhelming.💜

The Qwen-Image-Edit-2509 introduces multi-image editing support, allowing combinations like “person + person,” “person + product,” and “person + scene” with optimal performance using 1 to 3 input images. It enhances consistency in person, product, and text editing, preserving facial identity, product features, and text styles with improved precision. Additionally, it features native ControlNet support, including depth maps, edge maps, and keypoint maps, for more controlled and accurate edits. Try it here, make sure you click edit image below the text holder.

A woman in a black dress posing with a cigarette. Three women in light green dresses standing and gesturing with peace signs. Geometric patterns in blue, purple, and gold. A man in sunglasses and a scarf. An elderly man with glasses and a red scarf. A sports car on a street. A woman in a black dress standing by a car. An anime character with dark hair. Text in Japanese reading "株式会社" and additional text.

Luma AI just launched Ray3, built for production with high-fidelity HDR. It delivers state-of-the-art realism, physics, simulations, consistency, and adherence to prompts.

Love what you are seeing? Subscribe now! 😉

Hailuo’s First and End Frame feature, usually we want similar images for the best results, but I pushed the limits: a plain piece of paper as the start and a kirigami pangolin as the end. Didn’t expect it to turn out this wild!

I found this beautiful purple vibe, Ultraviolet style in Mage Space. You can create unlimted images for free here. Simple prompts make wonder with this style: A man is levitating in the sky, a lady is walking up a stairs, motion blur. Try it and share your creations with me by tagging me on X, Pinterest or Instagram. Or you can drop them in the chat here on Substack!

Neta Lumina, a fine-tuned Lumina model that crafts stunning manga-style and ACG art! I’m obsessed with its Guofeng aesthetic, and it handles Chinese, English, and Japanese prompts for epic creations! Why I love it?
1. Multilingual Power: Prompts in Chinese, English, or Japanese? Neta Lumina nails it.
2. Guofeng & ACG Mastery: From vibrant Guofeng ink-painting vibes to shonen, shojo, and Furry, it delivers crisp details and bold colors.
3. Complex Prompts Slayed: Try “1girl, hanfu, misty bamboo grove, by Tetsuya Nomura” for breathtaking art with multiple elements like character, attire, and setting.
4. Free & Open-Source
Try it here.

GenAIntel is this awesome spot where you just drop in a prompt and watch it run across over 40 AI video models all at once, side-by-side comparisons on an infinite canvas. It’s a total lifesaver for creators or video artists who want to quickly spot the best model without the hassle of switching tabs or subscriptions.

I’ve totally fallen down this rabbit hole, the Sref Explore page on Midjourney is where the magic happens for discovering insane style references. You can browse endless random codes, filter by vibes like color or mood, and even try them out live right there; no wonder I found the most stunning purple sref codes that make everything glow with that perfect ethereal hue.😉

I love to hear from you! Share your comments or questions below or reply to this email!

✨AI creative tools that you need now

Comments