Rad Teo

GPT 2.0 be like: What IP restrictions?? (I'll spare you from inappropriate versions...)

just made a short anime film🕊️ stack: seedance 2, mitte anime preset, vangogh

@GrimJim3 @EthanVanSciver Grok finally catch up with Kling 3.0 and Seedance 2.0

@EthanVanSciver Oh wow Grok is getting good...and here's me with a Friday off and a head full of ideas in support of CG think i might work on this other fan thing i wanted to try do! lol

This wasp is you.

This Midjourney V7 style reference feels like a fusion of classic European fantasy illustration and hand-drawn anime mythmaking: --sref 115126477 This style blends Franco-Belgian comic linework, Studio Ghibli-era character warmth, and mythological storybook painting. Delicate ink contours, airy painterly skies, classical Mediterranean palettes, and expressive character design give it the feel of an animated epic discovered in an old illustrated manuscript. Perfect for Greek mythology, gothic fairy tales, heroic legends, and poetic fantasy worlds.

I’m in the new Brittany Venti video briefly! Man, I’m just joking with my audience. People are listening tho!

Showing the same moment from multiple angles is a common film technique to amplify impact. I made an explosion version you can use as a base. You can easily adapt it to fight and crash scenes. Give it to your LLM and generate your own variations. Seedance 2.0 Prompt below 👇

In the spotlight🔦: Olivia Wilde as Quorra🔥 ✨️Fun on the Grid✨️

Oh wow, I love this cast for Mortal Kombat! 😲

Just tested GPT image 2.0, and it’s very good so far! Who would win this?

Honestly… GPT-2 is really strong. We finally have a real challenger to Banana 2. PROMPT Create an epic poster showcasing the most iconic moments of Michael Jordan career. epic, cinematic, lens flare

GPT’s Images2 × Seedance 2.0 is an absolute game changer ✅ Prompt "Generate a storyboard using a 3×3 grid layout." 1️⃣ Start by combining your concept with the ✅ prompt 2️⃣ Extract a detailed prompt from the result in step 1 3️⃣ Use that image as a reference inside Seedance 4️⃣ Paste the prompt from step 2 into Seedance The limitation? Video flexibility isn’t great yet, so outputs can feel a bit generic. With multiple scenes, the camera work often lacks impact. The key is to experiment with fewer or adjusted grid layouts to get better results.

📢JUST ADDED! RUNWAY now has full shot length control for SEEDANCE 2. That's right! You no longer have to select 5, 10 and 15 second shot lengths you can chose any increment from 4 - 15 seconds. VERY useful when using scene extensions you can now use 14 second shots as inputs and not have to cap it at 5 or 10 seconds. @runwayml

ChatGPT Images 2.0 just dropped. People can't believe it's 100% AI, it's insanely good. 10 wild examples: 1. "360 equirectangular image"

🚨 In less than 24h… GPT-2 is already blowing up. And it’s deserved. Simple prompts → seriously strong results. PROMPT Create Goku from Dragon Ball Z showing all his Super Saiyan transformations from base form to ultimate form in one cinematic composition.

Since the last post did really well, I’m dropping the exact prompt I used 👇 GPT’s Images2 × Seedance 2.0 workflow is powerful when used right 1️⃣ Combine your idea with the ✅ prompt 2️⃣ Let GPT generate the storyboard image 3️⃣ Use that image as a reference in Seedance 4️⃣ Paste the refined prompt back into Seedance This is the base I used to get consistent results. Still not perfect — motion and camera variety can feel limited — but with the right prompt structure, the quality jumps a lot. Next step is pushing this beyond “template feel” into something more cinematic. Prompt ⬇️

Prompt: "Restore this old photo into professional portrait of DLSR - quality colour and detail, using an advanced upscaling algorithm comparable to the results from canon EOS R6 II. Ensure the restored the image looks natural, retains exact facial features, has great clarity......"

Nano Banana Prompt ⏬ "A dramatic cinematic composition of a King of Hearts playing card emerging into reality. The card is slightly tilted in a dark, moody environment with soft spotlight lighting. The surface of the card is cracked and breaking apart, with debris and stone fragments flying outward in slow motion. From inside the shattered card, a realistic royal king steps forward — a middle-aged man with a strong, intense expression, short beard, and piercing eyes. He wears an ornate golden crown with red gemstones and a richly embroidered red royal coat with intricate gold patterns and fur details. The king is reaching forward and pointing directly at the viewer, creating a powerful, immersive 3D effect. His body is partially emerging out of the card, blending fantasy and realism. The card features classic King of Hearts symbols in red, with a textured, slightly worn surface. The background is dark, minimal, and cinematic with subtle smoke and light rays enhancing depth. Ultra-detailed textures, hyper-realistic lighting, depth of field, motion particles, sharp focus on the king, dramatic shadows, high contrast, 8K resolution, photorealistic, cinematic poster style."

Step 1: Generate the base image with GPT-2 Step 2: Serve the image you just generated back to GPT-2 with this prompt: "Convert this scene into a 360 equirectangular image"

GPT Image 2 is such a pleasure to use. It’s incredible how well it handles text, even in Spanish. For infographics, it doesn’t get better than this. I’ll definitely be using it a lot with my clients. I’d been wanting to make something like this too: an AI hater action figure 😅

A clean character intro template for Seedance 2.0 Just tweak it to fit your needs. You can also plug in your own character references, just tag them in the character section. Prompt: 15s / multi-cut STYLE: gritty crime, high sat, sharp contrast, deep shadows, warm practicals, grain STAGE: smoky retro pool hall, amber table lights, dark corners MUSIC: gritty jazz-hop ~85 BPM, dusty drums + bass hits, stop on final cue strike CHARACTERS: HOLIDAY: female, energetic, chaotic KODA: male, calm, controlled DEV: male, stoic, imposing FYE: female, confident, playful KARMA: agile stray cat SEQ: 1) 24mm MS / HOLIDAY walk → freeze TEXT [center-left]: "HOLIDAY. No plan. Still works." 2) 50mm MCU / KODA step → freeze TEXT [lower-right]: "KODA. Speaks rarely. Ends things." 3) 85mm CU / DEV crack → freeze TEXT [lower-left]: "DEV. Patient. Until not." 4) 35mm MS / FYE toss → freeze TEXT [upper-right]: "FYE. Knows the rules. Ignores them." 5) 35mm low MS / KARMA jump → freeze TEXT [lower-left]: "KARMA. Not yours. Just here." 6) ECU / cue strike → freeze TEXT [center]: "TABLE 5" MUSIC: stop