Datasets
VideoGaming
E-Sports Competition
Tournament-grade and high-rank competitive gameplay where small input differences determine outcome. Match metadata, ranking tier, and outcome are paired with each session.
Hours6.2K+
Games20+
Tiers4+
Training Use Cases
✓Imitation learning from expert gameplay
✓Action-conditioned video generation and world models
✓Game-state and match outcome prediction
✓Behavioral cloning at high skill levels
Key Highlights
✓20+ major esports titles covered across FPS, MOBA, RTS, fighting, battle royale, and racing genres
✓4+ competition formats covered including tournament, ranked-ladder play, scrim, and exhibition
✓Per-frame player input captured alongside the video
✓Tournament-grade competition rather than casual or scrim play
Metadata Fields
durationLength of clip in HH:MM:SS
resolutionPixel dimensions (e.g., 1920x1080, 3840x2160)
frame_rateFrames per second (e.g., 24, 30, 60, 120)
contains_audioWhether the clip carries an audio track (boolean)
primary_categoryDominant content category assigned to a video
stylefps | moba | rts | fighting | battle_royale | racing | sports_sim
competition_tiertournament | high_rank_ladder | scrim | exhibition
game_titleGame name and version
match_outcomewin | loss | draw | incomplete