TikTok-10M: 10 Million Short-Form Video Posts for Content Analysis

TikTok-10M is a large-scale dataset containing 10 million short-form posts from TikTok, designed for video understanding, multimodal learning, and social media content analysis. The dataset was curated by larlarHF to bridge the gap between academic video datasets and actual user-generated content, providing researchers with authentic patterns of modern short-form video. It was last updated on Hugging Face on March 24, 2026.

Use Cases

• Video understanding research based on authentic short-form video content.
• Multimodal learning experiments based on video and associated metadata.
• Social media content analysis based on patterns and characteristics of user-generated posts.
• Benchmarking models for short-form video recommendation or classification.

Strengths

• Contains 10 million data points, providing a large-scale resource.
• Focuses on authentic user-generated content from a major social media platform.
• Explicitly designed to bridge a gap between academic and real-world video data.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Freshness should be verified as the last update date is in the future (2026-03-24).

Provenance

Source

TikTok (user-generated content).

Collection Method

Curated from the platform; specific collection method is not detailed.

Freshness

Last updated 2026-03-24 07:40:45; freshness should be verified.