# ManiBench: A Benchmark for Testing Visual-Logic Drift and Syntactic Hallucinations in Manim Code Generation

arXiv: https://arxiv.org/abs/2603.13251 · HTML: https://mlsanity.com/paper/2603.13251/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 16 references that are in the map.
- First posted on arXiv 24 Feb 2026.

## Builds on (most cited first, 9 of 9)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Measuring Coding Challenge Competence With APPS](https://mlsanity.com/paper/2105.09938/) (2021)
- [DS-1000: A Natural and Reliable Benchmark for Data Science Code Generation](https://mlsanity.com/paper/2211.11501/) (2022)
- [SWE-bench Multimodal: Do AI Systems Generalize to Visual Software Domains?](https://mlsanity.com/paper/2410.03859/) (2024)
- [LLM Hallucinations in Practical Code Generation: Phenomena, Mechanism, and Mitigation](https://mlsanity.com/paper/2409.20550/) (2024)
- [HumanEval Pro and MBPP Pro: Evaluating Large Language Models on Self-invoking Code Generation](https://mlsanity.com/paper/2412.21199/) (2024)
- [Code2Video: A Code-centric Paradigm for Educational Video Generation](https://mlsanity.com/paper/2510.01174/) (2025)
- [SVGenius: Benchmarking LLMs in SVG Understanding, Editing and Generation](https://mlsanity.com/paper/2506.03139/) (2025)

## Built on by (most cited first, 2 of 2)

- [PRISM: A Benchmark for Programmatic Spatial-Temporal Reasoning](https://mlsanity.com/paper/2605.19382/) (2026)
- [See Before You Code: Learning Visual Priors for Spatially Aware Educational Animation Generation](https://mlsanity.com/paper/2605.15585/) (2026)

## Abstract (arXiv)

Traditional benchmarks like HumanEval and MBPP test logic and syntax effectively, but fail when code must produce dynamic, pedagogical visuals. We introduce ManiBench, a specialized benchmark evaluating LLM performance in generating Manim CE code, where temporal fidelity and version-aware API correctness are critical. ManiBench targets two key failure modes: Syntactic Hallucinations (valid Python referencing non-existent or deprecated Manim APIs) and Visual-Logic Drift (generated visuals diverging from intended mathematical logic through timing errors or missing causal relationships). The benchmark comprises 150-200 problems across five difficulty levels spanning calculus, linear algebra, probability, topology, and AI, grounded in analysis of 3Blue1Brown's ManimGL source (53,000 lines, 143 scene classes). Evaluation uses a four-tier framework measuring Executability, Version-Conflict Error Rate, Alignment Score, and Coverage Score. An open-source framework automates evaluation across multiple models and prompting strategies. Code, data and benchmark suite are available at https://github.com/nabin2004/ManiBench. and the dataset is hosted on https://huggingface.co/datasets/nabin2004/ManiBench.