VibePrevisBench introduces a benchmark for evaluating coding agents' ability to generate 4D scenes in Blender, featuring 242 tasks and two evaluation metrics.
On November 17, 2026, researchers led by Yuzhe Chen, Muxi Diao, and others introduced VibePrevisBench, a novel benchmark for evaluating coding agents' ability to generate 4D scenes in Blender. Published on arXiv (2611.15789) and highlighted on HuggingFace Daily Papers, the project aims to assess how effectively agents can translate briefs into executable 4D scenes and corresponding proxy videos.
VibePrevisBench evaluates coding agents across 242 tasks, categorized into four input types: scripts (94 tasks), photos (39 tasks), continuations (23 tasks), and reference videos (86 tasks). These tasks span seven scene domains and 33 locations, providing a comprehensive testbed for agentic coding capabilities.
In total, the benchmark incorporates 84,406 scorable hidden GT facts, ensuring rigorous evaluation of agent performance.
The research team analyzed the performance of 11 frontier coding agents, uncovering three key insights:
The VibePrevisBench project is fully open-source, with its code and data released on November 11, 2026. The GitHub repository (vibeprevisbench/vibeprevisbench) includes the complete evaluation pipeline, allowing users to test agents with a one-command evaluation tool, vibeprevisbench-eval. Additionally, the project homepage features benchmark results, an interactive leaderboard, and 14 case studies.
By providing a robust framework for evaluating agentic coding in 4D scene generation, VibePrevisBench sets a new standard for advancing AI-driven video generation technologies.
Choose your reading language and confirm your subscription by email. Unsubscribe at any time.
Source link







