Curious Refuge
@CuriousRefuge • 22,160 subscribers
The World's First Home for AI Filmmakers
Shorts
Videos
So Meta has officially entered the AI image and video space, and we're excited to share our early test results. Muse Video isn't publicly available just yet, so we recreated Meta's demo videos in Seedance to see how they compare. Based on these previews, Muse is looking pretty strong. We'll have to wait until it's officially released to know how it really stacks up. Up until now, it felt like everyone was talking about Seedance for professional AI filmmaking. It's exciting to see another competitive model enter the space and give filmmakers more options.
Curious Refuge31,192 views • 9 days ago
1:01
Sensitive content
This media may contain sensitive content.
We made a #StarWars trailer in the style of #wesanderson hope you guys enjoy it!
Curious Refuge2,302,997 views • 3 years ago
1:30
Sensitive content
This media may contain sensitive content.
What if Wes Anderson directed The Lord of the Rings? We asked the community which video they want to see next and Lord of the Rings took the cake… or should we say Elven bread. We hope you enjoy this Midjourney to Middle-Earth. #LordOfTheRings #WesAnderson #MovieTrailer #LOTR
Curious Refuge1,863,167 views • 3 years ago
Hey friends, here is a Star Wars-Themed AI Movie Trailer. We hope you enjoy it! 🙂 We created a tutorial that teaches you how to get the most cinematic shots possible from PikaLabs just like these. Check it out on our youtube channel! #ai #aifilms #pikalabs #midjourney #aiart
Curious Refuge702,836 views • 2 years ago
Finally got our hands on Seedance 2.0 (through YouArt) and ran it head-to-head with Kling AI 3.0. Here’s what stood out: Seedance 2.0’s physics feel stronger right away. The movement just looks more natural and realistic, especially in bigger physical actions. It also handles continuity across multiple shots from a single prompt better than I expected, which makes scene building a lot easier. And on the audio side, the lip sync generally holds up better than Kling. But here’s the tradeoff: Both models struggle with fine details. Background faces and smaller elements still get a bit messy. Seedance currently outputs at 720p, while Kling 3.0 supports 1080p, so Kling has the edge in overall sharpness. Even with upscaling, Seedance doesn’t fully recover those tiny details. Verdict🏆: Seedance 2.0 is a real step forward, especially for motion realism, but it’s definitely not the “end of Hollywood.” Check out our latest tutorial where we show you how to get access to this new model and take a deep look at the quality of this new AI Video Generator:
Curious Refuge58,038 views • 4 months ago
With the latest Seedance 2.0 release, there’s a feature we think might be even more transformative than the base video model itself: Seedance Omni. Similar to Kling Omni, Luma Modify, and Runway Aleph, Seedance Omni lets you guide the AI and make targeted edits to an existing video clip. It supports up to 9 reference images, 3 video clips, and 3 audio clips, allowing it to synthesize multiple layers of creative direction. We tested it across a range of scenarios (full prompts in the 🧵) 1. Modify eye color 2. Change weather 3. Time travel effect 4. Character swap 5. Add a spaceship 6. Change asteroids to meatballs 7. Dragon emerging from the clouds Verdict🏆: Seedance Omni excels at physical video dynamics, large visual effects, and environmental transformations. Its main weakness is resolution and output quality (around 720p), which can introduce flickering and softness.
Curious Refuge52,501 views • 4 months ago
We came across a really interesting workflow that could help solve environmental consistency. It’s called the “burst frame” technique. You take a single image, prompt 20 quick shot variations using Seedance 2.0, and generate a multi-angle sequence from the same scene. Because everything comes from one source, the environment stays consistent across shots. It ends up working like a fast pre-vis tool. Explore angles, pick your favorites, and build from there. Shoutout to Kai Turner for sharing this method.
Curious Refuge22,114 views • 2 months ago
Been playing with a character consistency workflow again. This came from Nexora that starts with a single GPT Image 2.0 character sheet and turns it into a multi-shot cinematic sequence in Seedance 2.0. One thing I added to the workflow: I first comped the character, environment, and visual style together in Midjourney before bringing everything into GPT Image 2.0 to build the character sheet. I also pushed it a bit further by generating multiple characters, destinations, and art directions at the same time, then turning those into production-style character sheets and shot sequences. It's interesting how much consistency you can get from a single well-defined character board when the visual language, wardrobe, and environment are established upfront. Check out three different character + scenes here 👇
Curious Refuge12,209 views • 1 month ago
So Runway Aleph 2.0 just dropped, and it introduces a pretty interesting workflow for AI video footage editing. Rather than repeatedly prompting a video model and hoping it makes the change you want, you can edit a specific frame first, then use that frame to guide the video generation. For this test, I take an existing footage inside Runway: • Picked a frame from the original video • Used Nano Banana Pro to add a new effect/object • Brought that frame into Aleph 2.0 to generate the updated video It feels like a more controllable and potentially cheaper way to make targeted edits to existing footage than current AI video omni models available like Seedance and Gemini.
Curious Refuge11,332 views • 1 month ago
Finally got to test out Veo 3 and my word it is an insane tool. We just launched this weeks AI Film News, where we compare Veo3 to Kling 2.0 and Runway, and it’s safe to say Veo is in the running as the best. Check out this weeks AI Film News and let me know if you agree:
Curious Refuge75,104 views • 1 year ago
We just tested a lip sync workflow in Seedance 2.0… and it might actually compete with HeyGen and other avatar models 👀 You upload: – a reference image – a blacked out video with audio OR audio file – a prompt (camera motion + VO) The prompted VO is really important here to make sure the model doesn’t hallucinate or riff off of the original speaker verses. Feels like a legit breakthrough. Lip sync has been a pain point for a while, and this gets us a lot closer.
Curious Refuge19,265 views • 2 months ago
We came across a really interesting tool that fixes a pretty common problem inside AI video workflows: extending AI cinematic scenes with seamless continuity. Using Omni models, you can take the last frame from an existing video clip and prompt something like: “show me what happens next in the video, use the last frame as the start frame” and it actually creates a second clip with surprisingly good continuity from the original shot. But you know how when you try stitching two clips together in Premiere Pro, you can usually spot the exact transition point? The exposure shifts slightly, the lighting bounces, or something just feels off enough that you can tell the clips don’t fully belong together. Well this new tool called SeeDance 2 Stitcher from TomLikesRobots🤖 automatically blends the two clips together in a much more seamless way and honestly the results are pretty impressive.
Curious Refuge14,083 views • 1 month ago
Can AI bring animation to life like Pixar? ✨ We took this Pixar-style Midjourney image and ran it through all 6 of our favorite AI video tools (Kling, Runway Gen-3, Pika Labs, Skyreels, Minimax and Luma) to see which could create the most magical motion. The clear champion? Pika Labs absolutely knocked it out of the park! The animation quality is jaw-dropping - seriously giving Pixar vibes with how smoothly it captured the character's wonder and the glowing flowers around her. Which animated AI videos have impressed you the most lately?
Curious Refuge62,341 views • 1 year ago
So Runway Gen 4.5 finally adds image-to-video, the workflow most pros rely on for consistency. We put it head to head with Kling AI and Flow by Google VEO using the same reference images and prompts (below) to evaluate motion quality, stability, and cinematic realism. 1. Action/WaterPhysics Test Prompt: Cinematic, wide-shot of a man running in a shallow river. The camera is tracking the man from behind as he runs up the river. Handheld camera shake as the camera follows the man. 2. Fire Physics Test Prompt: Cinematic, wide-shot of terrified woman running towards her burning barn. She abruptly stops, and puts in hands on her head as she watches her barn burn down. 3. VFX test prompt: Cinematic, wide-shot of a hooded figure. Flashes of purple magic and smoke whirl around the figure. The figure lifts its arms as the purple magic and smoke intensifies. 4. 2D Animation Test Prompt: 2D animated shot of a waiting at a bus stop in a thunderstorm. The man turns, walks to the bench, and sits down. 5. 3D Animation Test Prompt: 3D animated shot of an octopus. The octopus reaches into a coral and picks up a glowing white gem. 6. Conversation Test Prompt: slow camera push-in as two friends are having a conversation at a coffee shop Overall verdict: Despite the “world’s best” claim, Runway Gen 4.5 is not there yet. Prompt adherence is solid, but motion, physics, and cinematic realism still lag behind tools like Kling and VEO. Great platform, mid-tier model for now.
Curious Refuge25,538 views • 5 months ago
With the release of Kling AI 2.6, I was eager to put it up against one of its arch-rivals….Google Flow Veo 3.1. I ran four separate tests. First, I tested their ability to generate consistent and stable action shots. Second, I tested camera movement in a crowded scene. Third, I tested their ability to generate consistent facial details and emotion. Fourth, I wanted to see how well they can generated an interaction between two characters. Prompt 1: camera tracking shot, wide-shot. The camera tracks above as two nights ride on horses through a forest. The horses a galloping fast. The camera is tracking the knights from above. Prompt 2: Slow camera push in on a man eating Ramen at a Ramen house in Japan. The alley street is busy with people. Prompt 3: Close-up shot of a man on a sailboat. The man has a scared look on his face. After a short pause, the man says, "I'm gonna need a bigger boat” Prompt 4: Static camera. A man picks up a cookie from a plate and takes a bite. His dog looks up at him; eagerly wanting a bite of the cookie. The man turns, looks at the dog, and says, "no cookies for you mister” Verdict🏆: To be honest, I was surprised by Kling 2.6. While Veo 3.1 outperformed Kling 2.6 in the close-up shot test, it slightly underperformed in every other test. In the Veo 3.1 outputs, objects would randomly appear and camera movements were abrupt or didn’t adhere to the prompt. Don’t get me wrong, Kling 2.6 definitely had its flaws (its audio was quieter and often mismatched the prompt). That said, I was impressed with Kling 2.6 more than I imagined I would be. Although a somewhat small lead, I would give Kling 2.6 the slight advantage for this round of tests.
Curious Refuge24,695 views • 7 months ago
HUGE NEWS Hey friends! As you may have seen in The Hollywood Reporter today, Curious Refuge has been acquired by Promise to create the Hollywood studio (and school) of the future. We believe that technology has the ability to elevate voices from around the world and this partnership is the perfect way to connect artists (like you) with real-world opportunities. To celebrate, we’re launching The World’s First AI Animation Competition. With over $10,000 in prizes and a chance to pitch your film for real-world development, we’re super excited to see the incredible films you create. The winner will also receive their face printed on a potato which is super exciting… In light of all of this news we want to just reiterate that at its core Curious Refuge is, and will continue to be, a community of artists helping artists. In this new era we will continue to create and curate the best online AI courses, in-person meetups, competitions, content, helpful newsletters, and inspiration. We’re stoked to be on this incredible journey with you. Together, we want to help you share your creative voice, discover new opportunities, and find friends along the way. Because Curious Refuge isn’t just our name—it’s our promise to you. Much Love, Caleb & Shelby Read more here:
Curious Refuge42,378 views • 1 year ago
Check out our latest AI video comparison! This time we're putting Kling 1.6 head-to-head with Luma's new Ray2 image-to-video feature. We used the same image and the same prompt for both ("subtle camera movement, natural motion"), just like in our recent Pika 2.1 test. The results? Super interesting! Ray2 brings more dynamic motion and cinematic colors to the table, though with some distortion, while Kling keeps its reputation for solid quality and stability. Love seeing these different approaches to the same challenge. Which one do you like better?
Curious Refuge42,151 views • 1 year ago