Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Instead of prompting with text only, create reference sheets first. Prompts in the comments 👇👇 For a yoga / wellness influencer video, I build: First 15 seconds - Motion sheet - the exact routine, poses, timing, transitions Character sheet - same face, body, outfit, expressions Location sheet -studio, lighting,... show more

Dheepan Ratnam

14,809 subscribers

31,085 views • 3 months ago •via X (Twitter)

Health & Wellness Education

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

GPT Image 2 - Character Reference Sheet Prompt You can create a character reference sheet in any style you want for use in your Seedance 2.0 videos. All you need to do is either provide a reference image or describe your character. You can also input random values to generate completely random characters. This reference sheet provides a clear, structured, and highly readable visual breakdown of the character, helping the model maintain strong identity consistency across different outputs. The layout ensures the model focuses on essential visual information rather than stylistic noise, while the inclusion of expressions, head angles, and performance cues improves the model’s ability to generate believable acting and varied poses without losing character integrity. All clips in the demonstration video were generated using the reference sheet image and the same Seedance 2.0 prompt. You can find the prompts in the replies. 👇 GPT Image 2 and Seedance 2.0 are both available on mitte.ai

GPT Image 2 - Character Reference Sheet Prompt You can create a character reference sheet in any style you want for use in your Seedance 2.0 videos. All you need to do is either provide a reference image or describe your character. You can also input random values to generate completely random characters. This reference sheet provides a clear, structured, and highly readable visual breakdown of the character, helping the model maintain strong identity consistency across different outputs. The layout ensures the model focuses on essential visual information rather than stylistic noise, while the inclusion of expressions, head angles, and performance cues improves the model’s ability to generate believable acting and varied poses without losing character integrity. All clips in the demonstration video were generated using the reference sheet image and the same Seedance 2.0 prompt. You can find the prompts in the replies. 👇 GPT Image 2 and Seedance 2.0 are both available on mitte.ai

Kōda

162,620 views • 3 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 👇

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 Refuge

12,209 views • 1 month ago

Character LoRA on Flux – sneak peak 👀 I trained a model on 15 character images (same pose, different expressions) and now I can generate endless variants in new setups and poses. One image per inference, perfect consistency (both style + subject) with over 90% usable straight out of the box. Video below 👇 Trained on Scenario today.

Character LoRA on Flux – sneak peak 👀 I trained a model on 15 character images (same pose, different expressions) and now I can generate endless variants in new setups and poses. One image per inference, perfect consistency (both style + subject) with over 90% usable straight out of the box. Video below 👇 Trained on Scenario today.

Emm | scenario.com

48,818 views • 1 year ago

I stopped writing giant video prompts. Now I build the audio first with Seed Audio 1.0. Dialogue, pauses, emotion, sound cues, all in one generation. Then I feed that audio into Seedance and the video follows the rhythm on its own. Timing, pacing, even camera feeling, all coming from the sound. Audio first. Video second. This is a different way to make AI films.

I stopped writing giant video prompts. Now I build the audio first with Seed Audio 1.0. Dialogue, pauses, emotion, sound cues, all in one generation. Then I feed that audio into Seedance and the video follows the rhythm on its own. Timing, pacing, even camera feeling, all coming from the sound. Audio first. Video second. This is a different way to make AI films.

Anabiya

56,404 views • 1 month ago

google flow + claude + gpt image is actually f*cking cracked i figured out how to lock one AI character across an entire video. same face, same voice, same outfit, every single shot. zero drift. everyone making AI video right now hits the same wall. the character looks different in every clip so you can never tell an actual story. you get 5 second clips, not films. the fix isn't better prompting. it's building the character ONCE before you ever touch video, then referencing it everywhere. here's the workflow: > build the character inside Flow's character section, not a per-clip prompt > design the voice + personality one time and reuse it across every clip > generate a full turnaround sheet (Claude writes the prompt, GPT image renders the angles) > pull the character into any scene with a single @ tag > ingredients vs frames: when to let a shot float vs lock it to your source image > the end-frame loop that chains 4-10 sec clips into one seamless scene > stitch + trim in the timeline editor and export the whole thing as one film i packaged the entire thing. every prompt, the exact tool order, the character sheet template, and the continuity loop nobody actually explains. RT + reply "FLOW" and i'll send it over (must be following so i can dm)

google flow + claude + gpt image is actually f*cking cracked i figured out how to lock one AI character across an entire video. same face, same voice, same outfit, every single shot. zero drift. everyone making AI video right now hits the same wall. the character looks different in every clip so you can never tell an actual story. you get 5 second clips, not films. the fix isn't better prompting. it's building the character ONCE before you ever touch video, then referencing it everywhere. here's the workflow: > build the character inside Flow's character section, not a per-clip prompt > design the voice + personality one time and reuse it across every clip > generate a full turnaround sheet (Claude writes the prompt, GPT image renders the angles) > pull the character into any scene with a single @ tag > ingredients vs frames: when to let a shot float vs lock it to your source image > the end-frame loop that chains 4-10 sec clips into one seamless scene > stitch + trim in the timeline editor and export the whole thing as one film i packaged the entire thing. every prompt, the exact tool order, the character sheet template, and the continuity loop nobody actually explains. RT + reply "FLOW" and i'll send it over (must be following so i can dm)

Sulfur

35,486 views • 1 month ago

Create stunning cinematic videos using Kling 3.0 Omni, character sheets and multi-shot prompts. Here's how ... Kling 3.0 Omni can take up to 7 reference images. Working with a character sheet and passing it to Kling along with the environment and multishot prompts, allows you to quickly put cinematic videos together. Example prompt: cinematic. action movie. VFX. shot 1: close up of the hooded character. It's night. his eyes glow. he hears wolves in the distance and looks around and starts running. shot 2: full body shot of the character running. he looks over his shoulders. shot 3: we see a wolf running through the forest. it's hungry. shot 4:the character is stopped by a pack of wolves. shot 5: he fights the wolves with his stick. Link to Spaces below

Create stunning cinematic videos using Kling 3.0 Omni, character sheets and multi-shot prompts. Here's how ... Kling 3.0 Omni can take up to 7 reference images. Working with a character sheet and passing it to Kling along with the environment and multishot prompts, allows you to quickly put cinematic videos together. Example prompt: cinematic. action movie. VFX. shot 1: close up of the hooded character. It's night. his eyes glow. he hears wolves in the distance and looks around and starts running. shot 2: full body shot of the character running. he looks over his shoulders. shot 3: we see a wolf running through the forest. it's hungry. shot 4:the character is stopped by a pack of wolves. shot 5: he fights the wolves with his stick. Link to Spaces below

Martin LeBlanc

21,474 views • 5 months ago

🚨 Rockstar spends over $1,000,000,000 building GTA 6 a phone repair guy in Guangzhou shoots live-action GTA trailers with AI and cleared $12,684 last month. > Sheets: log 40 faceless channels, mark the clip beating its own median 30x: 1 day > Claude: pull 5 repeatable shapes out of the 20 winners, rewrite one into a 5-shot trailer: 20 min > CapCut: generate the shots back to back, caption them word by word: 40 min > Make: push to 3 platforms, write the new score back into the same sheet: auto > $54/month total. 2 days from the sheet to the upload. Rockstar needs a studio floor, 10 years and a motion-capture stage for one street. he needs a laptop and a text box. full research loop in the article above👇

🚨 Rockstar spends over $1,000,000,000 building GTA 6 a phone repair guy in Guangzhou shoots live-action GTA trailers with AI and cleared $12,684 last month. > Sheets: log 40 faceless channels, mark the clip beating its own median 30x: 1 day > Claude: pull 5 repeatable shapes out of the 20 winners, rewrite one into a 5-shot trailer: 20 min > CapCut: generate the shots back to back, caption them word by word: 40 min > Make: push to 3 platforms, write the new score back into the same sheet: auto > $54/month total. 2 days from the sheet to the upload. Rockstar needs a studio floor, 10 years and a motion-capture stage for one street. he needs a laptop and a text box. full research loop in the article above👇

Fokki

18,916 views • 6 days ago

The simplest way to create an AI influencer using workflow (3 steps): Nano Banana → Veo 3 → Arcads AI Workflow Step 1: Build your character with a JSON prompt (Nano Banana Workflow) → Use a structured JSON prompt instead of writing random text → Define the look clearly: character, outfit, lighting, camera style, background → Add negative constraints like: no logos, no text, no extra fingers, no CGI skin, no warped mirror This makes the output more stable and reusable. You’re not just prompting. You’re building a repeatable image system Step 2: Turn image → video (Veo 3 Workflow) → Connect the Nano Banana output directly into Veo 3 → Add a simple motion prompt: slight body movement, natural head turn, subtle phone movement → Keep it minimal so it feels more real Workflow removes the manual back-and-forth. Step 3: Add script + voice (Arcads AI Workflow) → Feed the generated video into Arcads AI → Add a short script, usually 10–20 seconds → Generate voice + lipsync that matches the character vibe Now the character becomes actual content. Why workflow matters: Prompt → Image → Video → Voice No messy manual process. No repeating the same setup every time. Just a clean pipeline you can scale. Nano Banana + Veo 3 + Arcads AI + workflow feature = simple AI influencer factory

The simplest way to create an AI influencer using workflow (3 steps): Nano Banana → Veo 3 → Arcads AI Workflow Step 1: Build your character with a JSON prompt (Nano Banana Workflow) → Use a structured JSON prompt instead of writing random text → Define the look clearly: character, outfit, lighting, camera style, background → Add negative constraints like: no logos, no text, no extra fingers, no CGI skin, no warped mirror This makes the output more stable and reusable. You’re not just prompting. You’re building a repeatable image system Step 2: Turn image → video (Veo 3 Workflow) → Connect the Nano Banana output directly into Veo 3 → Add a simple motion prompt: slight body movement, natural head turn, subtle phone movement → Keep it minimal so it feels more real Workflow removes the manual back-and-forth. Step 3: Add script + voice (Arcads AI Workflow) → Feed the generated video into Arcads AI → Add a short script, usually 10–20 seconds → Generate voice + lipsync that matches the character vibe Now the character becomes actual content. Why workflow matters: Prompt → Image → Video → Voice No messy manual process. No repeating the same setup every time. Just a clean pipeline you can scale. Nano Banana + Veo 3 + Arcads AI + workflow feature = simple AI influencer factory

Alex Nguyen

35,608 views • 4 months ago

A 19-euro tool just replaced 8,000 Meta employees. Here’s the exact workflow a TikTok creator used to build AI influencer videos from scratch — no camera, no team, no budget. Step 1. Open Pinterest. Search “Girl Self.” Save the image you like. Open Nanobanana Pro. Step 2. Swap the face using a single prompt. (Link at the bottom.) Step 3. Take your generated photo into Crust. Currently the easiest video-from-image tool on the market. Step 4. Create 3 nodes: reference image, text, video generator. Step 5. Drop your photo into the image node. Write a video prompt into the text field. This is what makes the photo move. Step 6. Connect the nodes. Text to text-in. Image to image-in. Select Kling 2.6. Set duration. Toggle sound. Step 7. Hit generate. 2 prompts. 1 AI girl. 0 employees. Meta spent $14B on Reality Labs last year to build digital humans. This creator built one in 7 minutes for less than a dinner. The prompts are free in the reply.

A 19-euro tool just replaced 8,000 Meta employees. Here’s the exact workflow a TikTok creator used to build AI influencer videos from scratch — no camera, no team, no budget. Step 1. Open Pinterest. Search “Girl Self.” Save the image you like. Open Nanobanana Pro. Step 2. Swap the face using a single prompt. (Link at the bottom.) Step 3. Take your generated photo into Crust. Currently the easiest video-from-image tool on the market. Step 4. Create 3 nodes: reference image, text, video generator. Step 5. Drop your photo into the image node. Write a video prompt into the text field. This is what makes the photo move. Step 6. Connect the nodes. Text to text-in. Image to image-in. Select Kling 2.6. Set duration. Toggle sound. Step 7. Hit generate. 2 prompts. 1 AI girl. 0 employees. Meta spent $14B on Reality Labs last year to build digital humans. This creator built one in 7 minutes for less than a dinner. The prompts are free in the reply.

RGK

186,635 views • 2 months ago

This is probably the most complex workflow I’ve ever built, only with open-source tools. It took my 4 days. It takes four inputs: author, title, and style; and generates a full visual animated story in one click in ComfyUI . I worked on it for four days. There are still some bugs, but here’s the first preview. Here’s a quick breakdown: - The four inputs are sent to LLMs with precise instructions to generate: first, prompts for images and image modifications; second, prompts for animations; third, prompts for generating music. - All voices are generated from the text and timed precisely, as they determine the length of each animation segment. - The first image and video are generated to serve as the title, but also as the guide for all other images created for the video. - Titles and subtitles are also added automatically in Comfy. - I also developed a lot of custom nodes for minor frame calculations, mostly to match audio and video. - The full system is a large loop that, for each line of text, generates an image and then a video from that image. The loop was the hardest part to build in this workflow, so it can process either a 20-second video or a 2-minute video with the same input. - There are multiple combinations of LLMs that try to understand the text in the best way to provide the best prompts for images and video. - The final video is assembled entirely within ComfyUI. - The music is generated based on the LLM output and matches the exact timing of the full animation. - Done! For reference, this workflow uses a lot of models and only works on an RTX 6000 Pro with plenty of RAM. My goal is not to replace humans, as I’ll try to explain later, this workflow is highly controlled and can be adapted or reworked at any point by real artists! My aim was to create a tool that can animate text in one go, allowing the AI some freedom while keeping a strict flow. I don’t know yet how I’ll share this workflow with people, I still need to polish it properly, but maybe through Patreon. Anyway, I hope you enjoy my research, and let’s always keep pushing further! :)

This is probably the most complex workflow I’ve ever built, only with open-source tools. It took my 4 days. It takes four inputs: author, title, and style; and generates a full visual animated story in one click in ComfyUI . I worked on it for four days. There are still some bugs, but here’s the first preview. Here’s a quick breakdown: - The four inputs are sent to LLMs with precise instructions to generate: first, prompts for images and image modifications; second, prompts for animations; third, prompts for generating music. - All voices are generated from the text and timed precisely, as they determine the length of each animation segment. - The first image and video are generated to serve as the title, but also as the guide for all other images created for the video. - Titles and subtitles are also added automatically in Comfy. - I also developed a lot of custom nodes for minor frame calculations, mostly to match audio and video. - The full system is a large loop that, for each line of text, generates an image and then a video from that image. The loop was the hardest part to build in this workflow, so it can process either a 20-second video or a 2-minute video with the same input. - There are multiple combinations of LLMs that try to understand the text in the best way to provide the best prompts for images and video. - The final video is assembled entirely within ComfyUI. - The music is generated based on the LLM output and matches the exact timing of the full animation. - Done! For reference, this workflow uses a lot of models and only works on an RTX 6000 Pro with plenty of RAM. My goal is not to replace humans, as I’ll try to explain later, this workflow is highly controlled and can be adapted or reworked at any point by real artists! My aim was to create a tool that can animate text in one go, allowing the AI some freedom while keeping a strict flow. I don’t know yet how I’ll share this workflow with people, I still need to polish it properly, but maybe through Patreon. Anyway, I hope you enjoy my research, and let’s always keep pushing further! :)

Lovis Odin

58,841 views • 10 months ago

I spent 5 hours today making OpenClaw even better at Motion Graphics I created the video below in 2 prompts with no asset uploads. It scraped or generated everything from a text prompt. I've added: - better brand scraping - image generation - video generation - music generation and scraping - better transitions on first prompt - basically zero errors - Always outputs a link to the editor you see below so I can open it from any device... (I can create and edit from telegram) Still work to do I will make a video this weekend and release the skill. It's the best remotion skill that exists imo.

Riley Brown

120,252 views • 5 months ago

🚨Marc Andreessen explains that the world has been s**tposting forever: "Newspapers have been scandal sheets forever. The first newspaper was a scandal sheet about The Vatican. It was all about scandals with the pope and the bishops, etc. Jefferson and Adams both owned newspapers and would use it to smear each other. Ben Franklin had 15 different sock puppet anon accounts (pseudonyms) and have them argue with each other in his newspaper. We've been in a world of information warfare for a very long time."

🚨Marc Andreessen explains that the world has been s**tposting forever: "Newspapers have been scandal sheets forever. The first newspaper was a scandal sheet about The Vatican. It was all about scandals with the pope and the bishops, etc. Jefferson and Adams both owned newspapers and would use it to smear each other. Ben Franklin had 15 different sock puppet anon accounts (pseudonyms) and have them argue with each other in his newspaper. We've been in a world of information warfare for a very long time."

Autism Capital 🧩

6,591,190 views • 1 year ago

// The Harmony of Pose-to-Pose and Layered Methods Explained (Continued) 1. In my previous post, I demonstrated the evolution of a standing roundhouse kick starting with Pose-to-Pose and transitioning into the Layered Method. 2. for this Hulk animation, I've opted for a different presentation format to ensure absolute clarity in showcasing the process. 3. the first 4 steps are how I animate 99.99999% of the time whether I'm working on AAA Games, Films, mobile or TV shows // Step #1 - I pose my character using 5 key poses (1 & 5 are duplicates. // Step #2 - I set the timing by moving ALL KEYS on the timeline until my eyes see something that works // Step #3 - I start adjusting my timing & F-curves. // Step #4 - I start using the Layered Method drop any questions you have in the comments 👇, I will answer them when I have a moment and a ☕️ in hand! ;)

// The Harmony of Pose-to-Pose and Layered Methods Explained (Continued) 1. In my previous post, I demonstrated the evolution of a standing roundhouse kick starting with Pose-to-Pose and transitioning into the Layered Method. 2. for this Hulk animation, I've opted for a different presentation format to ensure absolute clarity in showcasing the process. 3. the first 4 steps are how I animate 99.99999% of the time whether I'm working on AAA Games, Films, mobile or TV shows // Step #1 - I pose my character using 5 key poses (1 & 5 are duplicates. // Step #2 - I set the timing by moving ALL KEYS on the timeline until my eyes see something that works // Step #3 - I start adjusting my timing & F-curves. // Step #4 - I start using the Layered Method drop any questions you have in the comments 👇, I will answer them when I have a moment and a ☕️ in hand! ;)

John Animates

18,873 views • 2 years ago

FIRST TEST Seedance 2.0! From my initial tests, this is one of the more impressive AI video models I’ve tried so far. Dynamic motion feels fluid, prompt adherence is solid, and the efficiency really stood out, very little iteration needed. Everything you see here was generated using just one reference image + one prompt. I also tested a 3×3 image grid with no prompt at all, and the results stayed surprisingly consistent. Another highlight: sound design is already included, and it actually works really well. Tested via BytePlus 👉 More detail and access link to try Seedance 2.0 is in the comments.

FIRST TEST Seedance 2.0! From my initial tests, this is one of the more impressive AI video models I’ve tried so far. Dynamic motion feels fluid, prompt adherence is solid, and the efficiency really stood out, very little iteration needed. Everything you see here was generated using just one reference image + one prompt. I also tested a 3×3 image grid with no prompt at all, and the results stayed surprisingly consistent. Another highlight: sound design is already included, and it actually works really well. Tested via BytePlus 👉 More detail and access link to try Seedance 2.0 is in the comments.

Dinda Prasetyo

45,450 views • 6 months ago

GPT-5.6 Sol just changed how I build websites with AI. In this full 10-minute video, I test Sol and Terra inside Codex to build a cinematic landing page from a video reference. I cover: → Which GPT-5.6 model to use → How I analyze a full website experience → How I recreate the layout, motion and interactions → Why better prompting matters more than one-shot results → How to avoid generic AI slop The model matters—but the way you guide it matters even more. Watch the full process below 👇 Want the free template and the exact prompt I used? They’re available exclusively in the description of the YouTube version:

GPT-5.6 Sol just changed how I build websites with AI. In this full 10-minute video, I test Sol and Terra inside Codex to build a cinematic landing page from a video reference. I cover: → Which GPT-5.6 model to use → How I analyze a full website experience → How I recreate the layout, motion and interactions → Why better prompting matters more than one-shot results → How to avoid generic AI slop The model matters—but the way you guide it matters even more. Watch the full process below 👇 Want the free template and the exact prompt I used? They’re available exclusively in the description of the YouTube version:

Sourasith

261,047 views • 29 days ago

This guy got Claude to take control of Arcads and generate hyperrealistic UGC ads with a consistent AI actor, from a 7-word prompt, and now e-com brands pay him $1,500 per video to clone the setup. He got tired of watching DTC brands burn $4K on a photoshoot that dies the second a product angle changes, so he built a 2-prompt workflow that turns one product photo into scroll-stopping UGC without a single model, camera, or casting call. Here's the exact breakdown: → Claude is the director, Arcads is the studio. You never write a real prompt, you write one sentence → Type 7 words: "create a character sheet of a UGC actress on Arcads." Claude expands it into a 40-line spec: warm skin tone, high ponytail, gold hoops, white AirPods, athletic build, then takes control of Arcads itself → Arcads renders a 6-angle character sheet: front, back, profile, 3/4, two portraits, one consistent face locked across all of them → You download the grid and drop it back into Claude with one product photo → Second prompt, 8 words: "use Seedance 2.0 on Arcads to create natural UGC" → Claude writes the full shot list: timecodes, "1:03 AM clock zoom," the trackpad scroll, the arm pull that drags the product into frame, "delivery: casual, energetic," "ends on a wink + air kiss" → Arcads cuts the yellow Vans print onto an actor who was wearing all black in the grid. Real fabric folds, real light, zero warping → Final render: lip-sync locked to consonants, head not glued to the camera, a girl filming at 1am on her bed The key move 96% of people skip: you can't put the product on the actor before you lock the character sheet. If you drop a product photo onto a face that isn't locked across angles, the identity drifts frame to frame. The jaw resets. The skin tone shifts. The print slides off the fabric. The whole thing screams "AI" and your CTR dies in the first 2 seconds. His system locks the 6-view sheet first, so Arcads has one real human to dress, and the top lands with natural folds even though the grid never had it. One apparel brand ran 20 variants off a single product photo and found a winner in 36 hours, because the clips read like organic UGC, not a studio ad. Brands now pay him $1,500 per avatar build + $340/month to keep dropping fresh video batches on every new product. The entire thing runs on two prompts and one laptop. No photographer. No model agency. No product samples on set. Just two sentences, one product photo, and the discipline to lock the character before you render a single second of movement.

Shade

15,227 views • 2 months ago

Your one-minute clock starts now. Prompt 1 to Grok: "Review the video in the post. Create a detailed project requirements document to exactly recreate the app in the video" - about 15 seconds of typing Prompt 2 to Claude Code: "Go into planning mode and create a detailed plan to implement the app described in the attached PRD" - about 15 seconds of typing. Although I did have to come back and type "1" to give it permission to execute the plan. Prompt 3 to Claude Code: "Push this to a public repo." - about 5 seconds of typing All coding about 60 seconds of my time. The longest part was trying to figure out how to screen capture a video with system audio. That flummoxed me for three to four minutes.

Your one-minute clock starts now. Prompt 1 to Grok: "Review the video in the post. Create a detailed project requirements document to exactly recreate the app in the video" - about 15 seconds of typing Prompt 2 to Claude Code: "Go into planning mode and create a detailed plan to implement the app described in the attached PRD" - about 15 seconds of typing. Although I did have to come back and type "1" to give it permission to execute the plan. Prompt 3 to Claude Code: "Push this to a public repo." - about 5 seconds of typing All coding about 60 seconds of my time. The longest part was trying to figure out how to screen capture a video with system audio. That flummoxed me for three to four minutes.

Wayne Culbreth

615,311 views • 4 months ago

1 AI character clip becomes 50 ads without a single actor or reshoot. 100% generated. The face isn’t real. The pipeline is. Here’s the build, start to finish. Part 1 - Make the character. Generate the face on a node canvas. Lock the identity so it holds across every shot. Part 2 - Drop in the product. One node. The product sits in her hands, on the desk, wherever you point it. Part 3 - Animate. Seedance 2.0 turns the still into a moving clip. Lighting, motion, room - all baked in. Part 4 - The swap nodes. This is the part nobody shows. Build nodes that replace one thing at a time. Character. Outfit. Product. Nothing else moves. Swap the product → new ad. Swap the outfit → new ad. Swap the face → new ad. Same 6 seconds. Same light. Same motion. 50 variants before lunch. Old way: hire talent, film, reshoot for every SKU, edit each cut. New way: 1 canvas, swap 1 node, export. The cost of variant 2 used to be a full shoot. Now it’s a click. Film once. Swap forever.

1 AI character clip becomes 50 ads without a single actor or reshoot. 100% generated. The face isn’t real. The pipeline is. Here’s the build, start to finish. Part 1 - Make the character. Generate the face on a node canvas. Lock the identity so it holds across every shot. Part 2 - Drop in the product. One node. The product sits in her hands, on the desk, wherever you point it. Part 3 - Animate. Seedance 2.0 turns the still into a moving clip. Lighting, motion, room - all baked in. Part 4 - The swap nodes. This is the part nobody shows. Build nodes that replace one thing at a time. Character. Outfit. Product. Nothing else moves. Swap the product → new ad. Swap the outfit → new ad. Swap the face → new ad. Same 6 seconds. Same light. Same motion. 50 variants before lunch. Old way: hire talent, film, reshoot for every SKU, edit each cut. New way: 1 canvas, swap 1 node, export. The cost of variant 2 used to be a full shoot. Now it’s a click. Film once. Swap forever.

HodlReaper

21,188 views • 1 month ago

Here's how we made a 1-minute sequence of "The long road back". An animation short-film we've done at Magnific Studios. The process was roughly: - We started with reference images of the boy (Nilo), dog (Pip), the witch and witch house. Krea 2 image model, Nana Banana Pro were heavily used in reference image generation. - The video model was Seedance 2 generations. 15 seconds and with prompts asking for different shots. We basically prompted for 'scenes' - not shots. Trying to generate stand-alone shots and then editing together gives too many problems with continuity. - For every additional 15 second video (we needed at least 4), we would provide the last 5 seconds of the previous video as reference to Seedance 2 + the original reference images. Then in the prompt, ask for continuing the video + prompting for the next part of the story. This proved very productive. - We fixed inconsistencies in many shots by surgically extract the few problematic sections. Then edit 1 frame using Nano Banana. And then regenerating the sequences using this new reference image and Seedance 2. Two examples. The doorknob changed place and the door behind the dog was closed even after the witch stepped out. - All music was generated using Elevenlabs music generator. 3 seperate pieces of music. - All editing and color grading was done in Davinci Resolve. Everything gen AI above was done using Magnific

Here's how we made a 1-minute sequence of "The long road back". An animation short-film we've done at Magnific Studios. The process was roughly: - We started with reference images of the boy (Nilo), dog (Pip), the witch and witch house. Krea 2 image model, Nana Banana Pro were heavily used in reference image generation. - The video model was Seedance 2 generations. 15 seconds and with prompts asking for different shots. We basically prompted for 'scenes' - not shots. Trying to generate stand-alone shots and then editing together gives too many problems with continuity. - For every additional 15 second video (we needed at least 4), we would provide the last 5 seconds of the previous video as reference to Seedance 2 + the original reference images. Then in the prompt, ask for continuing the video + prompting for the next part of the story. This proved very productive. - We fixed inconsistencies in many shots by surgically extract the few problematic sections. Then edit 1 frame using Nano Banana. And then regenerating the sequences using this new reference image and Seedance 2. Two examples. The doorknob changed place and the door behind the dog was closed even after the witch stepped out. - All music was generated using Elevenlabs music generator. 3 seperate pieces of music. - All editing and color grading was done in Davinci Resolve. Everything gen AI above was done using Magnific

Martin LeBlanc

25,108 views • 1 month ago

Tamanna Bhatia 🩷 Created this by using a movement sheet as a reference image to animate the dance using Seedance 2.0 + ChatGPT image 2.0 GPT Image 2.0 Prompt: Dance Sequence Instruction Sheet [VISUAL STYLE] A composition featuring a highly detailed 3D-rendered female dancer. Designed like a professional choreography guide with a technical, diagram-inspired layout. Clean white background, soft studio lighting, and strong contrast to highlight body movement and posture. [GRID LAYOUT] Structured 4×4 panel grid (16 frames total), evenly spaced with thin black divider lines. Each panel is identical in size and clearly numbered from 1 to 16 to show a continuous dance progression. [CHARACTER] Use image1 as the base character. The same female dancer appears consistently across all panels with accurate likeness and proportions. [WARDROBE] The dancer wears a stylish, performance-ready outfit: a well-fitted top paired with a short, flowy skirt. The look should feel modern and visually appealing while still practical for dance movement. Fabric should subtly respond to motion (slight flow and folds), even in grayscale. [PANEL STRUCTURE – EACH FRAME] Top-left: Step number + short dance move title (e.g., “Step 5 – Spin Transition”) Center: Full-body pose capturing a precise moment in the choreography Bottom-left: 3–4 lines of concise instruction describing the move Overlay: Motion arrows and directional guides illustrating how the dancer transitions [MOTION INDICATORS] Incorporate curved arrows for fluid motion, straight arrows for directional steps, and circular indicators for spins or turns. Emphasize rhythm, weight shifts, and body isolation. [RENDER QUALITY] High-detail sculpted 3D style with smooth grayscale shading, subtle shadows, and clean linework. Maintain a polished, concept-art level finish with clarity in every pose. [RESTRICTIONS] No color, no background scenery, no extra characters, no visual clutter, only the dancer and instructional elements..

Tamanna Bhatia 🩷 Created this by using a movement sheet as a reference image to animate the dance using Seedance 2.0 + ChatGPT image 2.0 GPT Image 2.0 Prompt: Dance Sequence Instruction Sheet [VISUAL STYLE] A composition featuring a highly detailed 3D-rendered female dancer. Designed like a professional choreography guide with a technical, diagram-inspired layout. Clean white background, soft studio lighting, and strong contrast to highlight body movement and posture. [GRID LAYOUT] Structured 4×4 panel grid (16 frames total), evenly spaced with thin black divider lines. Each panel is identical in size and clearly numbered from 1 to 16 to show a continuous dance progression. [CHARACTER] Use image1 as the base character. The same female dancer appears consistently across all panels with accurate likeness and proportions. [WARDROBE] The dancer wears a stylish, performance-ready outfit: a well-fitted top paired with a short, flowy skirt. The look should feel modern and visually appealing while still practical for dance movement. Fabric should subtly respond to motion (slight flow and folds), even in grayscale. [PANEL STRUCTURE – EACH FRAME] Top-left: Step number + short dance move title (e.g., “Step 5 – Spin Transition”) Center: Full-body pose capturing a precise moment in the choreography Bottom-left: 3–4 lines of concise instruction describing the move Overlay: Motion arrows and directional guides illustrating how the dancer transitions [MOTION INDICATORS] Incorporate curved arrows for fluid motion, straight arrows for directional steps, and circular indicators for spins or turns. Emphasize rhythm, weight shifts, and body isolation. [RENDER QUALITY] High-detail sculpted 3D style with smooth grayscale shading, subtle shadows, and clean linework. Maintain a polished, concept-art level finish with clarity in every pose. [RESTRICTIONS] No color, no background scenery, no extra characters, no visual clutter, only the dancer and instructional elements..

Sydney

11,457 views • 3 months ago