Seedance 2.5 Test: Hyper-Realistic Video Generation
eyishazyer · x · 2026-08-14
A user shared a hands-on test and prompt for the video generation model Seedance 2.5.
The prompt specified generating a character from a reference image with exact facial features and hairstyle, creating a 30-second, 9:16 vertical late-night gym video. The results show exceptional handling of realistic skin textures (sweat, flushing) and raw iPhone footage aesthetics, praised as 'wildly realistic, genuinely feeling like raw gym footage.'
More from Multimodal
- Kling 4.0 fully available, shows markedly better motion coherence in fast-action scenes — alifcoder · 2026-10-16
- MJev-Spatial: Single-Pass Classification Beats Autoregressive VLMs on Spatial Reasoning — mindlab-research · 2026-10-16
- H-DDCM: Hadamard Codebooks Make Diffusion-Based Image and Video Compression Scalable — Technion · 2026-10-16
- ComfyUI v0.41.0 ships with Flux3 image model and MiniMax-H3 LightVAE support — Fcking_Chuck · 2026-10-16
- AI-generated short film 'The Dragon Rider' glides cinematically above the clouds — jijo6686 · 2026-10-16
- Realistic candid Midjourney portrait prompt: natural skin, lived-in bedroom light — tisch_eins · 2026-10-16