AI beyond manual labeling
now supercharged by foundation models

Accelerate time to value with our transformative approach to data-centric AI—powered by programmatic labeling and now, foundation models.

Request a demo

Data-centric AI technology developed at the Stanford AI Lab and proven at world-leading companies

Introducing Data-centric Foundation Model Development

New Snorkel Flow capabilities for enterprises to unlock complex, performance-critical use cases with GPT-3, RoBERTa, T5, and other foundation models.

Pixability distilled knowledge from foundation models and built smaller classification models with 90% accuracy in days with Data-centric Foundation Model Development in Snorkel Flow.

Learn more

The data-centric AI platform

Bring high-quality models to production faster with Snorkel Flow’s data-centric AI workflow. Accelerate AI development with programmatic labeling, rapid training data and model iteration, enhanced domain expert collaboration, and breakthrough application of all available knowledge—including the latest foundation models.

Automate data labeling

Use labeling functions (LFs) to programmatically label and encode rationale for labeling decisions. Partner with domain experts to curate data and troubleshoot errors. Snorkel Flow applies weak supervision techniques to intelligently combine and denoise LFs to auto-label vast training datasets in minutes.

Top US bank labeled 250K loan documents programmatically in under 24 hours.

See how it works

Train and analyze models

Guide rapid iteration with integrated model training and analysis to measure and improve training data and model quality. Choose from a library of leading models—including the latest foundation models—and train with a single click.

Memorial Sloan Kettering achieved 93% accuracy with rapid training data and model iteration.

See how it works

Leverage foundation models

Build large, domain-specific training sets to fine-tune foundation models in minutes. Or distill domain-specific knowledge from foundation models to train smaller, specialized deployable models. Snorkel Flow makes it possible to adopt foundation models for complex, performance-critical use cases.

Pixability distilled knowledge from foundation models and built a smaller deployment model with 90% accuracy in days.

See how it works

Get more from domain expert partnership

Don't limit your subject matter experts' participation to tediously labeling one by one. Empower them to transfer their knowledge in a fraction of the time required by manual labeling. Activate their expertise for dramatically better training data creation, iteration, and troubleshooting.

Global custodian bank saved 10,000 hours of investment managers' time by improving collaboration.

See how it works

Enterprise proven and ready

Cloud-agnostic and fully interoperable with your existing ML stack. Snorkel Flow provides enterprise-grade security and governance, user-tailored workflows, and access to unparalleled expertise.

Explore enterprise features

Data ingest

Quickly and securely integrate to data pipelines or upload data locally.

Model training

Train custom models or choose from leading model frameworks with optional AutoML.

Production serving

Deploy your models within Snorkel Flow or export to the service of your choice.

Infrastructure

Host Snorkel Flow within the secure infrastructure of your choice.

Deep platform support for complex ML tasks and data types

Build powerful AI solutions to handle real-world tasks over complex data types. Combine ML tasks for multi-model applications across a range of complex data types and formats.

Document classification

Information extraction

Structured data classification

Sequence tagging

Conversation analysis

Supported data types

Conversational text

Text documents

Native PDFs

HTML files

Semi-structured/ tabular data

Numeric data

Network data

And more

Accelerate AI development by 10-100x

Snorkel AI accelerates AI development for Fortune 500 enterprises, government agencies, and other AI innovators from months to minutes.

With Snorkel Flow, our teams can collaborate on data accuracy and improve time efficiencies for our highly valued physicians and medical professionals.

Janet Mak
Deputy CIO and VP of Digital Solutions

Learn more

Azure AI integrates Snorkel Flow's improved development cycle to automatically label and easily train new custom AI models.

Xuedong Huang
CTO, Azure AI

Learn more

Iterating training data and models took months before. It literally takes minutes using Snorkel Flow.

Swaroop Kalasapur
Head of Science Technology Innovation Center

Learn more

Snorkel Flow made the [clinical trial analytics] pipeline development a drag-and-drop experience.

Michael DAndrea
Principal Data Scientist

Learn more

70+ research publications on data-centric AI

Snorkel AI started as a research project at the Stanford AI Lab in 2015. Our founding team has spent over half a decade exploring programmatic labeling, weak supervision, and other data-centric AI techniques resulting in the Snorkel Research Project and 70+ peer-reviewed publications. We continue to invest in research with our academic partners, frequently publish in prominent journals, and productionize new advances in Snorkel Flow.

Learn more