Loading the SOTA2 catalog…
Compare training data & labeling products for capabilities, deployment, pricing, and enterprise readiness. Ranked using verified practitioner feedback, product research, and enterprise trust signals.
| Company | Description | Matching products |
|---|---|---|
| Sahara AI | AI workforce platform providing agents, data services, and developer tools for building, training, and deploying AI on-chain. | |
| Hub.xyz | Multimodal data lab that captures, curates, and annotates egocentric data for frontier AI and robotics companies. | |
| Scale AI provides data, evaluations, and AI systems for AI labs, governments, and the Fortune 500, operating across the AI stack from training data to deployed applications. |
| H2O.ai | Enterprise machine learning and AI platform company providing an end-to-end GenAI platform for predictive and generative AI on private, protected data. |
| Mercor | Mercor organizes human intelligence to power the AI economy, providing RLHF data, frontier research, and AI agent training at scale for top AI labs and enterprises. |
| Daz 3D | Daz 3D is dedicated to consistently developing technology for quick, realistic, and stunning 3D image rendering. |
| Snorkel AI | Builds specialized training data, benchmarks, and evaluation environments that help frontier models and agents perform in high-stakes domains. |
| Prolific | Prolific helps AI developers, researchers, and organizations easily access the highest-quality human data. Founded in 2014, it is used by 10,000+ organizations globally. |
| Roboflow | Computer vision platform for developers and enterprises to build, train, and deploy vision models. |
| Datacurve | Datacurve operates as a data engine for frontier AI, focusing on custom data for long-horizon reasoning, software engineering, and data science. |
| Surge AI | Surge AI is a data and AI company whose mission is to raise AGI with the richness of human intelligence — curious, witty, imaginative, and full of unexpected brilliance. Legally incorporated as Surge Labs, Inc. |
| Braintrust | The AI observability platform for tracing production, running evals, and catching regressions before they reach users. |
| LILT | Enterprise AI translation platform company providing multilingual content, translation workflows, and AI training data services across 100+ languages. |
| LanceDB | The multimodal lakehouse for AI, accelerating large-scale data curation and feature engineering so teams can build better models faster. |
| Labelbox | The RL data engine for AI teams. Post-training data, evals, and expert human intelligence at scale. |
| Human Archive | Human Archive is a data lab modeling human embodied intelligence. The company collects, labels, and synchronizes multimodal data including video, sensor, audio, and long-horizon inputs through a network of contributors and industry partners to create foundational infrastructure for robotics, spatial computing, and AI. |
| AfterQuery | Applied research lab curating data solutions to accelerate foundation model development. |
| ParallelDots | ParallelDots Inc. provides AI-powered retail execution and shelf analytics solutions for CPG and retail brands. |
| LLM Stats | AI & LLM leaderboard and comparison platform that ranks 300+ AI models by intelligence, speed, and price using a composite LLM Stats Score. |
| Sieve | The multimodal data lab. Provides high-quality video, audio, image, and interaction data for frontier AI. |
| Encord | The multimodal data layer for physical AI. Manage, curate, annotate, and align petabytes of data - from sensor streams to video to text. |
| HumanSignal, Inc. | Operates Label Studio, an open source data labeling and AI evaluation platform for multi-modal data including agent traces, LLM evals, RLHF, computer vision, document AI, NLP, and audio transcription. |
| BioSkryb Genomics | BioSkryb Genomics develops single-cell multiomic analysis products and bioinformatics platforms, including the BaseJumper Bioinformatics Platform and ResolveDNA amplification kits. |
| Lightly AG | Lightly AG provides a computer vision suite for machine learning teams, helping them label data, curate vision datasets, and pretrain models. |
| Segmed | Segmed provides de-identified medical imaging data for AI and clinical research, offering millions of regulatory-grade medical images to life sciences and technology companies. |
| Traverse | A research lab that partners with frontier AI labs to produce training data and reinforcement learning environments for models developing taste, judgment, and long-horizon work. |
| Klavis AI | Provides live environments for training AI agents, delivering high-quality coding and agentic data for frontier AI post-training, including long-horizon coding tasks, tool-use workflows, programmatic verification, granular rewards, and Dockerized environments for RL and SFT. |
| David AI | David AI is an audio data research company that develops proprietary audio datasets powering speech recognition, translation, synthesis, and conversational AI models for Fortune 100 companies and research labs. |
| Cortex AI | Builds real-world, real-workplace, and industry-scale egocentric and robot datasets for embodied AI. |
| Centaur Labs | Centaur Labs (d/b/a Centaur.ai) provides data annotation and model evaluation services using a global network of expert labelers and collective intelligence aggregation. |
| Robotics Center of Silicon Valley | A hub for showcasing, testing, and advancing robotics. Provides robot hardware sales and leasing, a robot learning data platform (Fearless), and a data marketplace. |
| Datasaur | AI lab building secure, private LLMs and agentic workflows for regulated enterprises, deploying custom AI solutions behind customer firewalls. |
| Besimple AI | Custom conversational audio datasets, expert annotation, and licensed voice actor data for training and evaluating speech and voice AI models. |
| Appen Limited | Provider of high-quality data for the AI lifecycle, offering AI training data, annotation, labeling, collection, and managed services across text, image, audio, video, and geospatial data. |
| Shofo Inc. | The World's Largest Video Library providing custom datasets from millions of hours of video, cleaned, segmented, and labeled for AI labs. |
| Velvet | A multimodal model R&D company that works with AI labs and enterprises to create datasets and evals advancing AI's interactive capabilities. |
| Nitrode | Provides high-quality spatial reasoning data that helps LLMs, agents, and world models understand dynamic environments, focused on advancing game development through frontier AI research. |
| Vibrant Labs | Applied research lab building novel data harvesting techniques, providing autoscaled RL data for frontier agents. |
| UBIAI | Provider of a text annotation tool and an LLM fine-tuning platform for building domain-specific AI models. |
| Ocular AI, Inc. | Applied research lab building data infrastructure and a human expertise network to encode real-world knowledge into frontier models across modalities, languages, and domains. |
| BioStack Platforms, Inc. | BioStack Platforms, Inc. provides differentiated data and training environments for healthcare AI models, including novel pre-clinical and medical datasets, causal inference, and multi-agent reasoning infrastructure. |
| Tenyks | Vision AI Platform and MLOps tools that empower businesses to achieve workforce efficiency, operational excellence, and better ML models through visual data intelligence. |
| Mundo AI | Mundo AI builds multimodal data infrastructure for research labs and Fortune 100 companies spanning audio, video, and emerging modalities to advance perceptual intelligence. |
| Vision Lab | The industrial data layer for robotics, providing egocentric footage from 50+ industries with SOP-level process knowledge. |
| Ooak Data | Ooak Data is an applied AI research lab that builds datasets, evaluation environments, and data pipelines, turning real company data into reinforcement learning environments for AI agents. |
| Sciloop | Expert-crafted evaluation and training data for frontier AI labs. A network of Olympiad medalists and top researchers across math, physics, chemistry, biology, and CS. |
| Panels | Provides high-quality audio datasets for training and evaluating speech and audio models, working with frontier labs and early-stage startups building at the edge of Voice AI. |
| Abundant | Builds environments and datasets for reinforcement learning (RL), powering top AI labs and enterprises with billions of tokens monthly. Based in San Francisco. |
| Agentic Labs | Provides a context engine for AI teams, collecting, cleaning, and organizing context for agent deployments. |
| Aluna | Data foundry for life science AI. Evals and datasets for life science AI. | DuplexDataset 1 more |