Compare vendors / Surge AI
Surge AI
Data major with off-the-shelf enterprise RL environments (EnterpriseBench)
Data majorApp replicasTool useCoding and SWEDomain experts
Surge’s EnterpriseBench suite includes CoreCraft, a customer-support world with 2,500+ entities and 23 tools; in its paper, one epoch of GRPO raised GLM 4.6 from 25.4% to 36.8% on held-out tasks. It has open-sourced the DAYJOB, HANDBOOK.md and GDP.xlsx harnesses under Apache-2.0 while keeping full task sets proprietary. Epoch AI names it among data providers now selling RL environments.
Best for: Enterprise customer-support and policy environments with a published training result
No outside funding disclosed (bootstrapped)
- – Bespoke builds
- – Platform or hub
- ✓ Open source
- – Publishes prices
Worth knowing before you sign
- Pending misclassification class action against its DataAnnotation arm (filed May 2025)
- Anthropic’s system card says it could not reproduce Surge’s GDP.pdf numbers (per rl-list)
- No public SOC 2 or trust page found
- Training results are vendor-run
Sources
- EnterpriseBench paper (arXiv)
- Surge blog: RL environments
- Surge homepage
- rl-list profile
- Bloomberg: funding talks
- Reuters via TipRanks: raise talks
- Bloomberg Law: misclassification suit
- Epoch AI: RL environments FAQ
- pavlovslist directory
Work at Surge AI? 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
- Vmax Generates new environments automatically from your data and evals