Verifiers

Prime Intellect's library for defining RL environments - a task dataset, a tool/sandbox harness, and a reward rubric - usable across training, eval, and synthetic-data workflows.

Open

Type: Framework
Publisher: Prime Intellect
Tags: RL Environments Reward Design
Runtime: verifiers
License: Apache-2.0
Size: Python library; ecosystem of 200+ Hub environments
Published: Jan 2025
Canonical: github.com/PrimeIntellect-ai/verifiers

Cite

Notes

Only stored in your browser.

Papers

introducesverifiers: Reinforcement Learning with LLMs in Verifiable Environments Introducing the Environments Hub verifiers: Reinforcement Learning with LLMs in Verifiable Environments

Contributors

Will Brown