USASI
SoftwareResearch stack

prime-rl

Project record

Maintained by Prime Intellect12

prime-rl is Prime Intellect's open-source framework for large-scale, asynchronous reinforcement learning and supervised fine-tuning of language models. It separates an FSDP2-based trainer, a vLLM inference service, and an orchestrator that gathers rollouts from verifiers environments, and it supports multi-node deployment on Slurm and Kubernetes.15

Last reviewedEntry updated Documented release Unknown

Availability and license

Overall availability

Public

Documented as available to the general public. Access conditions and license terms may still apply.12

Availability is separate from permission: read the license before using or redistributing.

The Apache 2.0 license covers the prime-rl repository. The verifiers library it integrates with is a separate Prime Intellect repository under the MIT License. This record does not assess the licenses of models or environments trained or used with prime-rl.23

Public materials checklist

Items for a research stack under USASI rubric v0.1. Unknown means unassessed or insufficient evidence.
Public materials checklist for prime-rl
ItemStatusNotes and evidence
Source codeIs the source code publicly readable?PublicPublic GitHub repository under the Apache License 2.0.12
DocumentationIs user documentation published?PublicThe repository's docs directory covers architecture, configuration, training, evaluation, scaling, algorithms, and development, and Prime Intellect's documentation site has a prime-rl section.14
Training codeDoes it include code for training models?PublicIncludes trainers for SFT and RL plus evaluation tooling, with optimized training code for selected model families.1
Data informationAre the data it expects or ships with documented?PublicThe README describes training on verifiers environments, installed from the Environments Hub or as opt-in workspace packages, and walks through example tasks such as text reversal, Wordle, and Wikipedia search.1
ReproducibilityAre instructions for reproducing reported results published?PartialThe README gives setup checks and end-to-end example runs, from small single-GPU tasks to Slurm-based training of larger models. A step-by-step reproduction of the INTELLECT-3 run was not found in the repository.1

What it is useful for

Post-training language models, including large mixture-of-experts models, with SFT, RL, and evaluations, from a single GPU up to large clusters. Prime Intellect used it to train INTELLECT-2 and INTELLECT-3.165

Run and use notes

Documented facts only. No hardware or performance claims are made without a cited source and stated assumptions.
  • The README states that at least one NVIDIA GPU is required, that the project is developed and tested on NVIDIA RTX 3090/4090/5090, A100, H100, H200, and B200 GPUs, and that setup uses uv with Python 3.12.1

Organization context

U.S. eligibility

Project eligibility rests on documented governing or maintaining entities, not on contributors.

Eligible · basis: U.S.-governed project

The repository is maintained in Prime Intellect's GitHub organization, its README cites Prime Intellect as author, and the INTELLECT-3 report introduces it as Prime Intellect's framework. Prime Intellect, Inc. is a Delaware corporation with U.S. locations (see the prime-intellect record).157

Assessed Sep 29, 2026

Sources

  1. 1.
    prime-rl: Async RL Training at Scale (README) (external site: github.com)

    Prime Intellect · Repository · accessed Sep 29, 2026

  2. 2.
    prime-rl LICENSE (Apache 2.0) (external site: raw.githubusercontent.com)

    Prime Intellect · License · accessed Sep 29, 2026

  3. 3.
    verifiers LICENSE (MIT) (external site: raw.githubusercontent.com)

    Prime Intellect · License · accessed Sep 29, 2026

  4. 4.
    Prime Intellect documentation (external site: docs.primeintellect.ai)

    Prime Intellect · Documentation · accessed Sep 29, 2026

  5. 5.
  6. 6.
    PrimeIntellect/INTELLECT-2 model card (external site: huggingface.co)

    Prime Intellect · Model card · accessed Sep 29, 2026

  7. 7.
    Prime Intellect, Inc. Form D (accession 0001231919-26-000720) (external site: sec.gov)

    U.S. Securities and Exchange Commission · Filing · published Jul 1, 2026 · accessed Sep 29, 2026

This listing is not an endorsement, a safety assessment, or a federal approval.

Support Us

Help keep USASI useful.

Optional. No USASI account required. Payment takes place on the linked provider’s website (Buy Me a Coffee).

About supporting this project