CLARU
memmelma

PEEK VQA

A dataset of 2M image-QA pairs for fine-tuning PEEK VLM, a vision language model for robotics that predicts trajectory paths and task-relevant masking points for robot manipulation. Answers are generated from Open X-Embodiment datasets using point-based representations normalized to [0,1]².

Downloads738

Technical Profile

Modalities
rgblanguage
Robot Embodiments
Franka PandaUR5JacoHydraEDANFanucStretch
Environment
labkitchen
Task Types
manipulationpick_and_place
Data Format
JSON
Part of the PEEK VQA family

Community Signals

Top 50% by downloads

Access

View on HuggingFace

Need custom rgb data?

Claru builds purpose-built datasets for lab applications with dense human annotations and quality assurance.

Request a Sample Pack

Related Datasets

LingBot-Depth Dataset

Self-curated RGB-D dataset for training masked depth modeling approaches, containing real-world indoor scenes, VLA robot manipulation tasks, and simulated data across multiple camera types and robot platforms.

rgbdepth
167K downloadsApr 2026CC BY-NC-SA 4.0

OmniAction

A large-scale multimodal dataset for proactive robot manipulation comprising 141,162 episodes across 112 skills and 748 objects, enriched with audio, visual, and contextual instruction data for cross-modal intention recognition.

rgbaudiolanguage
145K downloadsMar 2026cc-by-nc-4.0

Xperience-10M

A large-scale egocentric multimodal dataset of human experience containing 10 million interactions and 10,000 hours of synchronized first-person recordings with six video streams, audio, stereo depth, camera pose, hand mocap, full-body mocap, IMU, and hierarchical language annotations for embodied AI, robotics, and world modeling research.

rgbaudiodepthproprioception+3
127K downloadsApr 2026other

Egocentric-100K

The largest dataset of manual labor with 100,405 hours of egocentric video from head-mounted fisheye cameras, featuring state-of-the-art hand visibility and active manipulation density.

rgbvideo
74K downloadsFeb 2026apache-2.0

OmniAction

A large-scale multimodal dataset for proactive robot manipulation comprising 141,162 episodes with cross-modal contextual instructions derived from spoken dialogue, environmental sounds, and visual cues rather than explicit commands.

rgbaudiolanguage
71K downloadsMar 2026cc-by-nc-4.0

Open-H-Embodiment

A community-driven, multi-embodiment dataset of paired kinematics and video for training and evaluating AI autonomy models in surgical robotics and ultrasound applications, including tabletop exercises, clinical procedures, and healthcare robotics simulations.

rgbproprioception
63K downloadsMay 2026cc-by-4.0