Oxen Labs, Inc.
Oxen.ai aims to make versioning datasets as easy as versioning code, providing a platform for ML engineers to collaborate, fine-tune, and deploy models at scale.
At a Glance
- ML Engineering Teams
- Generative AI Startups
- Creative Studios
- Data Scientists
AI Tools by Oxen Labs, Inc.
(1)
Oxen.ai
AI Lifecycle Platform for Teams
Discussions
No discussions yet
Be the first to start a discussion about Oxen Labs, Inc.
Latest News
Oxen's Model Report: Weekly summary of new AI models and research
How a $1 Qwen3-VL Fine-Tune Beat Gemini 3
Cutting Inference Costs from $46K to $6.5K Fine-Tuning Qwen-Image-Edit
Guide: How to Set Noise Timesteps When Fine-Tuning Diffusion Models
Products & Services
End-to-end platform for versioning datasets, fine-tuning models, and serverless deployment.
Lightning-fast version control system specifically designed for large ML datasets and unstructured data.
High-performance generative image model available on the platform.
API for running inference on custom or community AI models without infrastructure management.
Market Position
Oxen.ai differentiates itself through extreme performance in handling large unstructured datasets and a familiar Git-like developer experience.
Leadership
Founders
Greg Schoeninger
CEO and Founder. Previously Head of Engineering at Fyter and Hologram Inc. Software Engineer at IBM and Alchemy API (acquired by IBM). Has been building AI/ML developer tools since 2012.
Scott Howard
Co-Founder and COO. Founder of House of Ventures. Previously Head of Operations at Share Ventures and Founder of Day Zero Ventures. Background in venture capital and operational strategy.
Executive Team
Greg Schoeninger
Chief Executive Officer
Expert in AI/ML infrastructure and developer tools. Former Head of Engineering at Fyter and Hologram Inc.
Scott Howard
Chief Operating Officer
Seasoned entrepreneur and investor with experience in scaling venture-backed companies.
Board of Directors
Founding Story
Founded by Greg Schoeninger, who previously worked at Alchemy API and IBM. The company was started to solve the friction of managing large datasets in AI development, creating a 'Git for data' that is optimized for scale and speed.
Business Model
Revenue Model
Tiered subscription model for repository hosting (Free, Hacker, Pro) plus pay-as-you-go GPU compute for fine-tuning and inference.
Pricing Tiers
Unlimited public repositories, 5 private repositories, 50 GB storage and transfer.
Unlimited private repositories, 100 GB storage and transfer.
Unlimited private repositories, 500 GB storage and transfer.
Target Markets
- ML Engineering Teams
- Generative AI Startups
- Creative Studios
- Data Scientists
- Custom model fine-tuning on proprietary data
- Generative media creation (images, video, audio)
- Collaborative dataset curation for ML teams
- Reproducible ML experimentation workflows
- Meta
- Dolby Labs
- JBL
- Samsung