VibeVideoBench: multimodal agents coding videos often fail on time management
Muxi Diao · hf · 2026-10-28
Researchers from BUPT, Zhongguancun Academy, PKU and the Kling Team introduce VibeVideoBench, a benchmark evaluating multimodal agents that generate videos via code. Tasks span games, concept explainers, build processes and paper explanations, testing systems like Kimi K3, DeepSeek-V4.1, Gemini 4 and GPT-6.
Key finding: top systems meet core requirements on most tasks but frequently fail due to poor time management. Project page: vibevideobench.github.io.
More from coding & agent
- Investor Gavin Baker: Grok Bot is the most impactful AI product at his firm, with 17 bots built — elonmusk · 2026-10-28
- Claude reads my social feeds nightly and shows highlights on an Apple TV dashboard — florianvo · 2026-10-28
- Developer slams Apple: each iOS simulator stores its own Siri model, wasting 25.5 GiB — Dimillian · 2026-10-28
- Dev demos mobile AI companion that voice-controls your Mac and orchestrates agents — RileyRalmuto · 2026-10-28
- Dev builds open-source Excalidraw MCP after AI-generated diagrams kept overlapping text and arrows — pikahikmag · 2026-10-28
- Amazon tells engineers to get customer-facing or lose their future as AI reshapes coding — DynamicWebPaige · 2026-10-28