---
format: "aidr-story-markdown/v1"
id: "324ea250080273cb9bc05bcf77dfe912c70ca4e8f1808b6ae243898720467d69"
canonical_url: "https://aidr.today/324ea250?lang=en"
title: "Code2Video-Compass: A Comprehensive Benchmark for General-Purpose Code-Based Video Generation across T2V, I2V, and V2V"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-23T00:00:00.000Z"
category: "Data"
topics: ["llm","data-engineering"]
source_urls: ["https://huggingface.co/papers/2610.28416"]
summary: "Researchers have introduced Code2Video-Compass, a general-purpose benchmark for code-based video generation covering text-to-video, image-to-video, and video-to-video settings. The benchmark includes 2,104 tasks across 10 categories and 24 subtypes, evaluated using 12 objective and five subjective metrics."
---

# Code2Video\-Compass: A Comprehensive Benchmark for General\-Purpose Code\-Based Video Generation across T2V, I2V, and V2V

> [Open the canonical story](<https://aidr.today/324ea250?lang=en>)

**Published:** 2026-10-23T00:00:00.000Z
**Category:** Data
**Topics:** llm, data\-engineering

## Summary

Researchers have introduced Code2Video\-Compass, a general\-purpose benchmark for code\-based video generation covering text\-to\-video, image\-to\-video, and video\-to\-video settings\. The benchmark includes 2,104 tasks across 10 categories and 24 subtypes, evaluated using 12 objective and five subjective metrics\.

## Sources

- [Story source](<https://huggingface.co/papers/2610.28416>)