# ActivityNet-QA

ActivityNet-QA is an AI benchmark; this page records what it measures and traces its data, grader, and scores to primary sources.

Tests question answering over long videos with questions that require spatial and temporal reasoning.

## Details

- Address: Online, Global
- Website: [github.com/MILVLG/activitynet-qa](<https://github.com/MILVLG/activitynet-qa>)
- Tags: Helpful, Science
- Category: Multimodal
- Status: Active
- Data licence: MIT
- Data access: open
- Grader: exact
- Organization: MILVLG
- Released: 2019-01
- Repository: [github.com/MILVLG/activitynet-qa](<https://github.com/MILVLG/activitynet-qa>)
- Task count: 58,000
- Top-score provenance: unknown
- ABI family: No

## Questions

### Where is ActivityNet-QA?

ActivityNet-QA is at Online, Global.

### How do I contact ActivityNet-QA?

To contact ActivityNet-QA, visit https://github.com/MILVLG/activitynet-qa.

## Source

From [Benchmark primary sources](https://aibenchmarkindex.com/about), license CC0-1.0, checked 2026-10-15.

Last updated October 15, 2026.

## Links

- This benchmark on the [web page](<https://aibenchmarkindex.com/global/online/activitynet-qa>)
- [Multimodal in Online](<https://aibenchmarkindex.com/global/online/category/multimodal.md>)
- [More benchmarks in Online](<https://aibenchmarkindex.com/global/online.md>)

> **Check before you compare.** Benchmark methods and leaderboards change. Follow the cited primary source before using a result in a decision.