Commercial training data from specialist providers. Listings show what each source advertises; licensing is confirmed with the provider before any agreement.
1. Image Description Data Data Size: 500 million pairs Image Type: generic scene(portrait, landscapes, animals,etc), human action, picture book, magazine, PPT&chart, App…
Overview This massive, globally sourced imagery dataset provides access to over 300 million high-quality visual assets.
A multi-turn trajectory dataset that captures the full process of Korean-speaking users and AI agents working together to complete tasks.
Dataset Size: 1,700 proprietary high-resolution 16-bit images.
A collection of multi‐sensor dataset collected across lithuania and wider eu road networks, featuring high‐resolution rgb video from front, rear, side, fisheye, and…
1. Overview the provider has off-the-shelf 1 million videos of multimodal data, including multi-style video, lip syn, emotion video and etc.
A single-turn Agent / CRM domain speech dataset based on the North African Arabic (Maghrebi Arabic) dialect.
The PPT Visual Rendering Dataset is a large-scale collection of 850K+ unique PowerPoint presentation files in PPT and PPTX format, built as AI training data for…
2,000 robot manipulation episodes with multi-camera video, robot joint states, gripper states, action trajectories, object labels, task phases, and success/failure…
The 100K+ Hours STEM Classroom Video Dataset is a comprehensive collection of classroom videos of 100K+ hours in Science, Technology, Engineering, and Mathematics (STEM)…
Dataset Description The 850K+ Scalable Vector Graphics (SVG) Image Dataset is a large-scale collection of high-quality, resolution-independent vector images designed for…
Swedish TTS dataset with spoken sentences read by native speakers.
A collection of visual design assets including infographics, posters, reports, presentations, newsletters, cards, etc along with respective json file generated using…
A collection of indoor and aerial environment scenes (342,389 lidar interior and 77,777 drone), captured with rgbd data, refined trajectories, and scene meshes.
A single-turn Agent / CRM domain speech dataset based on Modern Standard Arabic (MSA).