Devs: Sora makes indistinguishable video while image gen still looks like slop
Angaisb_ · x · 2026-11-09
gabriel1 argues Sora remains the only video model producing truly indistinguishable footage, but likens it to GPT-4o text: the first output looks human, ten outputs in it's clearly mode-collapsed slop. Angaisb agrees, wondering how Sora 2 can make perfect videos while OpenAI's image generation still looks like slop — wishing the Sora team built image gen instead.
Related event: Developer says Sora remains the only model producing truly realistic videos(2 posts)→
More from Models
- SuperGrok $100-tier user hits usage wall on a Qwen 27B query, threatens to cancel — gnukeith · 2026-11-09
- Internal benchmark chart fuels speculation Anthropic and OpenAI are a generation ahead internally — haider1 · 2026-11-09
- Opus 5.5 nails 3D modeling but fumbles a simple football icon that gpt-image-2 aces — rms80 · 2026-11-09
- Thinky Inkling 2 Small: New Open US Model Trails DeepSeek Flash While Costing More — intellectronica · 2026-11-09
- Gemini 4 Carbon Ranks #1 on Artificial Analysis but No One Can Actually Use It — Yuchenj_UW · 2026-11-09
- Users report inconsistent AI bot allowance usage and weekend performance dips — mattyp · 2026-11-09