we train all of our models until they "saturate" which usually happens around 1M steps using a very large batch size models memorize the same amount, regardless of training datasize meaning they have fixed capacity and instead "spread it thinner" when trained on more examples
GENERATIVE AI
-

GPT Model Capacity Analysis: Linear Trend at 3.6 Bits-Per-Parameter
By
–
we then compute the capacity of different models (GPT models with varying numbers of layers and hidden dimensions) averaged over hundreds of models in fp32, we get the following curve, indicating a linear trend of around 3.6 bits-per-parameter, regardless of the exact details:
-
AI Filmmaking Workflow and Feedback Opportunity
By
–
Pick one template, swap the nouns, run it, iterate 3×, and post the best 8 s cut. Drop your first AI-generated video below. I’ll give feedback. Let’s push the frontier of AI filmmaking together.
-
Prompt Engineering Tips for Fixing Common Generative AI Visual Artifacts
By
–
Common fails → quick fixes Jerky motion ⇒ add “cinematic 24 fps with motion interpolation.” Melting faces ⇒ include “stable identity, photoreal human, no distortion.” Lighting mismatch ⇒ “global HDR grade” or relight with Runway “Relight” filter.
-
AI-Powered Video Post-Production Workflow
By
–
Post-production flow 1. Generate A-roll clips (4–8 s each).
2. Export to DaVinci/Premiere or Capcut; slice at beat markers.
3. Add LUT + grain overlay for cohesion.
4. Upscale to 4 K with Topaz or Runway Super-Res.
5. Caption burn-in with your brand font. -
Advanced Prompt Engineering Techniques for AI Models
By
–
8. Advanced prompt tricks • Parentheses weighting: (highly detailed:1.4) boosts fidelity.
• Negative prompts: “no watermark, no distorted digits.”
• Rhyme schemes for audio: “Up-tempo hip-hop beat, 95 BPM, minor scale,” gets Veo 3’s music generator on point. -
Practical AI Editing and Remixing Workflows
By
–
7. Editing & remix hacks • Runway Motion Brush: draw where you want extra movement.
• Kling timeline: stitch multiple shots, automatic color match.
• Image-to-video: feed a keyframe into Sora/Kling for consistent characters across clips.
• Layer Veo 3 audio onto older silent -
Practical Prompting Techniques for Video Generation with Veo 3
By
–
6. Scene tweaking on the fly
— God of Prompt (@godofprompt) 3 juin 2025
Want night instead of day?
→ “Replace lighting: neon twilight, purple-blue hue.”
Need energy?
→ “Add dynamic handheld shake, subtle motion blur.”
Swap props?
→ “Change mushroom to floating lantern, preserve lighting response.”
Veo 3’s inline… pic.twitter.com/UW6nJIqp5b6. Scene tweaking on the fly Want night instead of day? → “Replace lighting: neon twilight, purple-blue hue.”
Need energy?
→ “Add dynamic handheld shake, subtle motion blur.”
Swap props?
→ “Change mushroom to floating lantern, preserve lighting response.” Veo 3’s inline -
Prompt Template for Pixar-Style AI Animation
By
–
Template: Pixar-style Animation "Whimsical forest clearing at dawn
Chubby robot watering glowing mushrooms, top-down crane shot
Soft pastel palette, depth-mapped volumetric fog
Playful marimba soundtrack, cinematic 24 fps" Add “toon shader, thick outlines” for comic vibes. -
A 4-part structure for precision AI image prompts
By
–
5. The 4-part precision prompt
— God of Prompt (@godofprompt) 3 juin 2025
[Scene] [Subject (+action)] [Camera/Motion] [Mood + Style]
Example →
[SCENE] Mist-covered Scottish highlands, sunrise
[SUBJECT + ACTION] Red stag steps onto cliff edge, antlers backlit
[CAMERA/MOTION] 300 mm telephoto push-in, gentle stabilised… pic.twitter.com/sUTIdGhCxz5. The 4-part precision prompt [Scene] [Subject (+action)] [Camera/Motion] [Mood + Style] Example → [SCENE] Mist-covered Scottish highlands, sunrise
[SUBJECT + ACTION] Red stag steps onto cliff edge, antlers backlit
[CAMERA/MOTION] 300 mm telephoto push-in, gentle stabilised
