Hugging Face· PlatformResearch paper

TimeScope: How Long Can Your Video Large Multimodal Model Go?

TimeScope is an open-source benchmark designed to measure how well vision-language models understand long videos, evaluating localized retrieval, information synthesis, and fine-grained temporal perception across videos from 1 minute to 8 hours.

From the source

TimeScope is an open-source benchmark designed to measure how well vision-language models understand long videos. By adding short “needle” clips into videos ranging from 1 minute to 8 hours, it evaluates three skills: localized retrieval, information synthesis, fine-grained temporal perception.

huggingface.co

Announced

23 Jul 2025 00:00 UTC

Detected

03 Aug 2026 21:12 UTC (+376d after)