Robotics AI training data providers.
44 companies say on their own websites that they sell datasets and work on robotics. 23 of them have revenue, a valuation or funding on record, and come first, largest first. Each line below is the company's own description, linked to the page it is on.
“High-quality training data, annotations, and RLHF to power the world's most advanced AI models.” its page
Data comes from: made by experts and contractors, replicas of real software.
“Pre-built AI training datasets for LLMs. Scalable, expert-curated data for model training, evaluation, and reasoning performance.” its page
Data comes from: made by experts and contractors, recorded from people and the real world.
“We helped a leading tech giant to build a diverse, egocentric dataset to solve this.” its page
Data comes from: made by experts and contractors, recorded from people and the real world.
“Appen provides specialist LLM training data across the full development pipeline, from initial fine-tuning demonstrations through to the expert preference feedback and adversarial testing that alignment requires.” its page
Data comes from: recorded from people and the real world, made by experts and contractors, licensed from its owners, generated.
“Egoverse is Mecka's large-scale human interaction dataset, built from real-world first-person activity across tasks, spaces, and environments” its page
Data comes from: recorded from people and the real world.
“Expert-built, verified by code, and proven to transfer. ... License our off-the-shelf data for eval, SFT or RL.” its page
Data comes from: recorded from people and the real world, made by experts and contractors, generated.
“A larger, privately held dataset, which we license to companies for their own internal validation.” its page
Data comes from: made by experts and contractors.
- Lightwheel
#22 overall, $145M raised in total
“The first batch of EgoSuite-Open100K is live on the Hugging Face Hub, open for academic research and commercial training, with the remainder rolling out in stages.” its page
Data comes from: recorded from people and the real world, made by experts and contractors, generated, public data.
“Training-ready data for physical AI, collected by in-field operators and at our teleoperation facilities.” its page
Data comes from: made by experts and contractors, recorded from people and the real world, generated.
“delivering the highest-quality teleoperation, on-embodiment, and egocentric data” its page
Data comes from: made by experts and contractors, recorded from people and the real world.
- CloudFactory
#26 overall, $65M raised in total
“We transform messy, unstructured, or incomplete data into high-quality datasets, creating the foundation for reliable AI models.” its page
Data comes from: made by experts and contractors.
- Defined.ai
#27 overall, $62.3M raised in total
“Access off-the-shelf ethically sourced and high-quality datasets across audio, image, video, text, and multimodal formats” its page
Data comes from: recorded from people and the real world, made by experts and contractors.
“Centific’s Data Marketplace provides a centralized, self-service platform to discover, enhance, and deliver trusted datasets from both Centific and third-party sources at scale.” its page
Data comes from: recorded from people and the real world, licensed from its owners, the client's own data, made by experts and contractors.
- SuperAnnotate
#30 overall, $49.5M raised in total
“Versioned, audited datasets with full lineage from raw signal to train set.” its page
Data comes from: made by experts and contractors.
“Explore, license, and analyze production-ready datasets or brief us on a custom build we scope from scratch.” its page
Data comes from: made by experts and contractors, recorded from people and the real world, licensed from its owners.
“Audio, video, visual, and doc intelligence datasets for mid/post training and evals.” its page
Data comes from: made by experts and contractors.
“Original corpus of 1,000 Chain of Thought math problems, including step-wise corrections and annotation, intersecting 60+ disparate subdomains of formal and applied mathematics.” its page
Data comes from: recorded from people and the real world, made by experts and contractors.
- Midcentury
#39 overall, $15M raised in total
“One of the largest unscripted egocentric datasets across industrial and everyday environments.” its page
Data comes from: recorded from people and the real world.
“Access continuously refreshed, expert-labeled multimodal data designed for teams building real-world robotics systems.” its page
Data comes from: made by experts and contractors, recorded from people and the real world.
- Human Archive
#44 overall, $8.2M raised in total
“Leading frontier AI labs trust us for their video, sensor, robot deployment, image, and audio datasets.” its page
Data comes from: recorded from people and the real world, made by experts and contractors.
- VisionLab AI
#47 overall, $6M raised in total
“Its datasets include both egocentric footage, showing how trained workers perform tasks, and exocentric footage, capturing broader factory context, workflows, tools, and operating conditions.” its page
Data comes from: recorded from people and the real world, made by experts and contractors.
“Egocentric-10K is the largest egocentric dataset. It is the first dataset collected exclusively in real factories.” its page
Data comes from: recorded from people and the real world.
“Curate from millions of hours of content and 100+ video, audio, and gaming datasets.” its page
Data comes from: licensed from its owners, the client's own data, recorded from people and the real world.
“Provide processed datasets and results to the 6thSense partners and customers those datasets are captured for.” its page
Data comes from: recorded from people and the real world.
“Datasets with repository patches, programming problems, and code comprehension tasks.” its page
Data comes from: made by experts and contractors, licensed from its owners, recorded from people and the real world.
“We collect human activity data at scale from real environments across multiple continents, and turn it into clean, annotated datasets that teach robots to understand the physical world.” its page
Data comes from: recorded from people and the real world.
“Explore 100+ licensed datasets across robotics, video, and multimodal AI.” its page
Data comes from: licensed from its owners, recorded from people and the real world, made by experts and contractors.
“Building the world's most diverse real-world, real-workplace, and industry-scale egocentric + robot dataset, accelerating our progress toward physical intelligence.” its page
Data comes from: recorded from people and the real world, made by experts and contractors.
“Cerebro is now the data refinery that turns raw operational data into training-grade datasets: SFT data, RL data, and RL environments graded on real outcomes.” its page
Data comes from: the client's own data, generated.
“Datoric develops custom datasets for robotics, treating research, collection, verification, and production as one continuous process.” its page
Data comes from: made by experts and contractors, recorded from people and the real world.
“Enact turns a policy’s failures into targeted recovery datasets for teams training robotics models.” its page
Data comes from: recorded from people and the real world.
“Humyn Labs builds custom AI training datasets across voice, image, video, and sensor modalities collected and annotated by verified domain experts with multi-layer quality control.” its page
Data comes from: made by experts and contractors, recorded from people and the real world.
“Large-scale conversational datasets across 22 official Indian languages.” its page
Data comes from: recorded from people and the real world.
“Power your robotic systems with high-fidelity spatial data and unique datasets for embodied AI created from scratch.” its page
Data comes from: made by experts and contractors, recorded from people and the real world, replicas of real software.
“Collect, annotate, and ship training-ready egocentric and multimodal robotics datasets with verified quality.” its page
Data comes from: recorded from people and the real world, made by experts and contractors.
“Share the behaviors, embodiment, and volume you need. Let's discuss the right dataset.” its page
Data comes from: recorded from people and the real world.
“Elite annotators. Scalable synthetic generation. One pipeline built for enterprise AI.” its page
Data comes from: made by experts and contractors, the client's own data, generated.
“Shaip’s AI Data Catalog & Licensing Marketplace gives AI teams a single source for buying and licensing pre-labeled, commercially cleared training datasets across text, speech, image, video, and multimodal formats.” its page
Data comes from: made by experts and contractors, recorded from people and the real world.
“We package training-ready datasets, evaluation sets, and environments for secure delivery.” its page
Data comes from: recorded from people and the real world, licensed from its owners.
“from building high-quality training datasets to delivering production-ready models” its page
Data comes from: the client's own data.
“Frontier AI training data, engineering and intelligent automation to accelerate responsible AI rooted in real ROI.” its page
Data comes from: not stated.
“We prepare ready-to-use and custom datasets supporting a range of AI use cases” its page
Data comes from: made by experts and contractors, recorded from people and the real world.
“High-quality labeled datasets with real-world data and synthetic samples for AI models, machine learning, and generative AI” its page
Data comes from: recorded from people and the real world, generated.
“Scripted and spontaneous recordings from diverse speaker populations.” its page
Data comes from: made by experts and contractors, recorded from people and the real world.
