Compare vendors / Vmax
Vmax
Generates new environments automatically from your data and evals
Platform or hubCoding and SWETool use
Vmax sells Campaign, a self-play task-generation product that turns a customer’s data and evals into new sets of environments that can be downloaded. Its research includes PROPEL, procedurally generated unix-ctf environments and GlyphBench (360+ game tasks). The founders hold RL PhDs, and South Park Commons and Race Capital are listed as backers.
Best for: Teams that want to expand an existing eval set into many more environments
1 Legal nameVmax
2 HeadquartersSan Francisco, CA
3 Founded2025
4 TeamAbout 11
5 DeliveryPlatform or hub, Bespoke builds
6 ExclusivityNot disclosed
7 Customers and partnersNot disclosed
8 Published pricesNot disclosed
Worth knowing before you sign
- Campaign is waitlist-only
- No named paying customers
- Funding undisclosed; no code linked from its papers
Sources
- Vmax homepage
- South Park Commons profile
- Martian: ARES collaboration
- PROPEL paper (arXiv)
- unix-ctf paper (arXiv)
- rl-list profile
- pavlovslist directory
Work at Vmax? Send corrections with a source to editors@toprlenvironments.com. We fix verified errors within five business days.
Compare with
- Prime Intellect Runs the Environments Hub, open-source RL tooling and hosted RL training
- HUD Platform and marketplace for building, buying and running RL environments
- General Reasoning Hosts RL environments as API endpoints through OpenReward
- Veris AI Simulated copies of a company’s systems for testing and training enterprise agents
- Chakra Labs Expert-authored tasks and cloned enterprise apps for training and red-teaming