INTELLECT-3.1

INTELLECT-3.1: A 100B+ MoE trained with large-scale RL

Trained with prime-rl and verifiers
Environments released on Environments Hub
Read the Blog & Technical Report
X | Discord | Prime Intellect Platform

Introduction

INTELLECT-3.1 is a 106B (A12B) parameter Mixture-of-Experts reasoning model built as a continued training of INTELLECT-3 with additional reinforcement learning on math, coding, software engineering, and agentic tasks.

Training was performed with prime-rl using environments built with the verifiers library. All training and evaluation environments are available on the Environments Hub.

The model, training frameworks, and environments are open-sourced under fully-permissive licenses (MIT and Apache 2.0).

For more details, see the technical report.

Serving with vLLM

The model can be served on 2x H200s:

vllm serve PrimeIntellect/INTELLECT-3.1 \
 --tensor-parallel-size 2 \
 --enable-auto-tool-choice \
 --tool-call-parser qwen3_coder \
 --reasoning-parser deepseek_r1

Citation

@misc{intellect3.1,
 title={INTELLECT-3.1: Technical Report},
 author={Prime Intellect Team},
 year={2025},
 url={https://huggingface.co/PrimeIntellect/INTELLECT-3.1}
}

Base model

zai-org/GLM-4.5-Air-Base

(6)

INTELLECT-3: A 100B+ MoE trained with large-scale RL • 5 items • Updated Feb 18 • 12