INTELLECT-3 is a 106B (A12B) parameter Mixture-of-Experts reasoning model post-trained from
GLM-4.5-Air-Base using supervised fine-tuning (SFT) followed by large-scale reinforcement learning (RL).
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.
INTELLECT-3 achieves best-in-class performance on math, coding, and reasoning benchmarks:
1vllm serve PrimeIntellect/INTELLECT-3 \
2 --tensor-parallel-size 2 \
3 --enable-auto-tool-choice \
4 --tool-call-parser qwen3_coder \
5 --reasoning-parser deepseek_r1
1vllm serve PrimeIntellect/INTELLECT-3-FP8 \
2 --enable-auto-tool-choice \
3 --tool-call-parser qwen3_coder \
4 --reasoning-parser deepseek_r1
1@misc{intellect3,
2 title={INTELLECT-3: Technical Report},
3 author={Prime Intellect Team},
4 year={2025},
5 url={https://huggingface.co/PrimeIntellect/INTELLECT-3}
6}
