@ChatGPT would love the ability to PIN a specific response within a chat so I can easily come back to it later. Would be especially helpful for long threads!
Been digging into emotional performance tests between @seedance2_ai and @Kling_ai lately and it’s pretty wild seeing how far AI video has come.
Some of the subtle acting and facial nuance genuinely surprised me.
Click through this post for the deep dive!
We tested Seedance 2.0 against Kling 3.0 across a range of emotional performances using naturalistic language in the prompt to see how each model handles cinematic acting.
Across most of the tests, Seedance felt more emotionally nuanced and natural, while Kling often leaned a bit broader and more theatrical in its performances.
We broke down all the tests in our latest blog post here:
https://t.co/CW7H0U7lHs
This AI parody of a dating reality TV show might even be more entertaining than the real thing. Adam Palmer nailed it, from the set and character design (that jawline 👀) to the twist that had us going “of cooourse he is” and down to the details… is that a GPU ring?!!! 😂
Exploring a soft pink blue hazy color palette here in this cinematic set.
Here’s the master prompt:
“dreamy cinematic wide shot of a [SUBJECT], soft ethereal glow, light leak effect, film burn highlights, overexposed sunlight, pastel pink and blue color wash, soft haze diffusion, glow bloom, low contrast, lifted blacks, subtle motion blur, shallow depth of field, nostalgic mood, 35mm film grain, shot on Kodak Portra 400, analog film texture, poetic, serene, surreal, soft focus”
If you try it out, share it with us! We'd love to see what you come up with 👀
Tried out a style reference from @midjourney and am really liking the dramatic teal and orange neo noir aesthetic here. I also put it through @Magnific_AI Skin Enhancer so the image feels more realistic and the skin appears less plasticky, though it does alter the original style a bit. Tradeoffs!
If you want to try this out all you need to do is write your prompt and then include: “-- sref 579989385”
“Still Here” is an ad about a rechargeable battery that lasts through relationships, jobs, and other life changes. But the battery becomes more than something that lasts. It becomes an emotional anchor, quietly carrying through different moments in life.
#RunwayBigAdContest
I created an AI short film — AURAE 🪭 , questioning the absence of female heroes in Asian mythological narratives.
As an Asian woman, I grew up with stories of male heroes—Sun Wukong, Nezha, Erlang Shen—deeply embedded in our collective memory. Yet my mother told me stories of Mulan, Mu Guiying, and Liu Sanjie—equally brave, yet far less visible. Over time, I realized that within widely circulated mythologies, female heroes are rare—almost absent.
So I brought this question onto the screen: to imagine a mythological heroine who is not flawless, but vulnerable, wounded, and searching—yet never defined by those moments. She can fall into darkness, but she can also choose to rise.
Because we are not pre-written characters—we choose who we become.
This short film was created in collaboration with LibTV by LibLib, an AI video platform for professional creators. Its “infinite canvas” interface visualizes the creative process in real time, enabling rapid iteration and adjustment. It shifts creation away from fragmented file management and closer to the flow of thinking itself.
Playing with pastel green and warm tones in this cinematic set. What story do you see unfolding through these stills?
Master Prompt here (and individual prompts in ALT):
"Cinematic still of a [SUBJECT] [DESTINATION], soft ethereal glow, dreamy haze diffusion, shallow depth of field, wind-blown hair, delicate flowing dress, natural environment, soft bloom highlights, lens flare, volumetric sunlight rays, shot on 35mm film, Kodak Portra 400 color grading, pastel greens and warm tones, low contrast, subtle film grain, nostalgic, poetic, serene"
If you try it out, share it with us! We'd love to see what you come up with 👀
Exploring retro halftone print textures in this set of cinematic stills, inspired by vintage film posters. Here’s the master prompt:
"cinematic still, [SUBJECT], soft focus background, halftone print texture, CMYK dot pattern, retro magazine print aesthetic, faded film colors, 35mm film grain"
If you try it out, share it with us. We'd love to see it!
Wow, love the energy of this ad made with AI. The retro 90s art direction and catchy beats make it super fun, and the edit is sick. Like, I am sold, get me one of those, please.
Well done, Karsten Winegeart @Karsten116 👏🌈🤩
I call this collection “Head in the Clouds.” ☁️
These cinematic stills were prompted in Midjourney using a moodboard. The aesthetic leans dreamy, hazy, nostalgic, and romantic, with huge skies and quiet human moments.
If you want to try this look yourself, add this moodboard code to the end of your prompt in Midjourney: --p mwrto64
Share your work with us, we’d love to see it!
Playing with Chiaroscuro-inspired lighting with dramatic, color-driven shadows. Made with Midjourney + Magnific skin-enhancing tool. What do you think?
Sharing the master prompt (see ALT for specific prompts):
Cinematic studio portrait of a [SUBJECT], minimalist high-fashion editorial style. Dramatic chiaroscuro lighting with [COLOR] light shaping [PART OF FACE], naturally falling off into deep shadow across the rest of the face, set against [BACKGROUND COLOR]. Moody noir lighting, extreme contrast, deep blacks, intentional negative space. Graphic lighting composition with sharp but natural falloff, no diffusion, sculpted facial planes barely emerging from darkness. Shot on 35mm film at f/2.8, cinematic depth of field, fine film grain, hyper-realistic skin texture with natural pores, restrained color grading, ultra-detailed, 8k.
If you try it, tag us! Would love to see what you create.
Playing with Y2K aesthetics in @midjourney + @Magnific_AI's Skin Enhancer. What do you think? What early-2000s deets would you add to the prompt?
Prompt: woman wearing a bright yellow fitted crop tank top with bold black graphic text, dark low-rise pants, white wraparound Y2K sunglasses, tight cornrow braids pulled straight back, minimal makeup, natural lips, three-quarter profile pose with head tilted downward, cool introspective expression, graphic editorial styling, color remapping, posterized color treatment, heavy film grain
Been experimenting with glass refraction and distortion as a subtle way to make things look more interesting. I love how it adds a layer of intrigue to these portraits.
Here’s the master prompt I used in @midjourney 👇
portrait of a [SUBJECT DESCRIPTION], photographed entirely through vertically reeded architectural glass with medium grooves running top to bottom across the entire frame, full-frame optical distortion dominating the image, realistic physical light refraction creating vertical repetitions across the subject, no unobstructed areas, [LIGHTING STYLE], [COLOR PALETTE / GRADING], [MOOD / EDITORIAL STYLE], physical optics not digital, no glitch --style raw
See ALT for the specific prompts:
When @BytedanceTalk first launched @seedanceai last year, it was one of the best AI video tools out there. Over time, it kind of took a back seat to models like VEO and Kling. Since Seedance 1.5 Pro added native audio, I’ve been curious how it stacks up against VEO 3.1 and Kling 2.6, so I put these to the test:
Prompt 1: a handheld shot of a woman yellling “I hate you.” with anger in her face
Prompt 2: Two boxers are facing off in a boxing ring. They’re warming up and getting ready to fight. The man on the left says, “get ready to lose again” to which the man on the right replies, “let’s dance buttercup”
Prompt 3: Three teenagers are standing at a faire. The two girls are eating cotton candy and the boy says, “did you know cotton candy isn’t made of cotton”. To which one of the girl replies, “really!? then what about elephant ears?”
Verdict🏆: Seedance 1.5 Pro actually performs really well alongside VEO and Kling, especially when it comes to sticking to the script. As always, iteration matters, but this definitely feels like a solid comeback for Seedance.
Which model is currently your favorite? 👀
Ancient soul meets next-gen visuals. ⚡️
The debut AIMV "WU AI-HUA" is officially out!
Warner Music China's newest virtual idol, AI-HUA@ Wuaihua, redefines the Wuxia spirit for a global audience. Created in collaboration with Kling AI Creator @ jcwooofficial, this project shatters the boundaries between tradition and modern beats.
Turn up the volume and step into the new frontier of digital storytelling. 🎧
I’ve been playing around with these whimsical, motion-blurred cinematic stills in Midjourney, where the subject is clear and the rest blurs around them.
Kind of feels like life right now… fast, busy, and still reminding us to slow down and enjoy the little things as we wrap up the year ❤️
Prompt: cinematic still of a woman standing with eyes closed, arms gently open as a chaotic street blurs behind her like she’s letting go, wide shot from slightly above, motion forming a halo effect around the subject, sunlit tones, soft wind movement, empowering mood
I call this Disco Euphoria. Made at a recent @MachineCinemaAI GenJam. Adam Mutchler's approach: "You have a better outcome if you ride the wave of AI, embrace it like an untamed beast, as opposed to controlling it," really inspired the work.
Really loving this moody, low-key glamorous aesthetic. The bokeh lights really give it a fun night-out energy. 🌃⚡Created in Midjourney. See ALT for prompts!
What vibe should we try next?