Datasets
VideoGaming
Multiplayer Interaction
Multi-agent gameplay where multiple human players interact in the same session, with each participant's inputs captured and synchronized to the shared video feed. Cooperative, adversarial, and free-form formats are all represented, with voice and text chat preserved as recorded.
Hours48.3K+
Games30+
Players2-32
Training Use Cases
✓Multi-agent reinforcement learning
✓Imitation learning from human teamplay
✓Action-conditioned world models for multiplayer environments
✓Game-state and intent prediction
Key Highlights
✓30+ game titles covered across cooperative, adversarial, and free-form multiplayer formats
✓2-to-32 player session sizes spanning duos, squads, and large-team matches
✓Per-player input streams synchronized with shared video for every session
✓Voice and text chat preserved alongside gameplay, not stripped from the recordings
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
stylephotorealistic | stylized | low_poly | retro_pixel
interaction_formatcooperative | adversarial | free_form | mixed
player_countNumber of human players in the session
game_titleGame name and version
communication_presentWhether voice or text communication is captured alongside (boolean)