verifiers
Prime Intellect’s MIT library for creating LLM reinforcement-learning environments and evals, with a matching PyPI package.
Tags
Description
Established Prime Intellect project (canonical repo PrimeIntellect-ai/verifiers; not archived; created 2025-01-22). Not presented as a brand-new 2026 launch. Root LICENSE is MIT (Copyright 2026 Prime Intellect). PyPI verifiers 0.4.1 re-verified 2026-11-13 (not yanked; wheel uploaded 2026-11-03; project URLs point at this repository). README describes it as a library for creating environments to train and evaluate LLMs, tightly integrated with Prime Intellect’s Environments Hub, the prime-rl trainer, and a Hosted Training platform. The Environments Hub and Hosted Training are Prime Intellect products (https://app.primeintellect.ai/…); they may require accounts and can incur real compute cost — no signup or spending was done. Environments can execute task or agent code. Repo docs/ and AGENTS.md are in-tree guides; no separate docs hostname was used. Same operator family as prime-rl.
Agent access
Related
- prime-rl: Prime Intellect’s Apache-2.0 asynchronous RL training framework, installed from git/uv (no verified PyPI package) and integrated with verifiers environments.
- OpenEnv: BSD-3-Clause framework for building and serving isolated agentic RL environments over a Gymnasium-style reset/step/state API.
- Harbor: Framework from the Terminal-Bench creators for evaluating and optimizing agents and models in sandboxed environments, run in parallel across cloud sandbox providers.
- NeMo Gym: NVIDIA Apache-2.0 library for building, evaluating and scaling agent environments, verifiers and harnesses, including sandboxed code-execution tasks.
Guides & comparisons
- RL environment libraries compared — NeMo Gym, OpenEnv and Verifiers (Comparison): Compare NeMo Gym, OpenEnv and Verifiers as libraries for writing and running RL environments: authoring model, isolation, documented trainers, install paths, licences and current versions. Distinct from Harbor/AgentEnv/Seahaven and from the agent RL stacks comparison.
Sources
- https://github.com/PrimeIntellect-ai/verifiers
- https://github.com/PrimeIntellect-ai/verifiers/blob/main/LICENSE
- https://github.com/PrimeIntellect-ai/verifiers/blob/main/README.md
- https://github.com/PrimeIntellect-ai/verifiers/tree/main/docs
- https://pypi.org/project/verifiers/
Machine-readable
- JSON
- Markdown
- Source file on GitHub (edit via pull request)