INTELLECT-3
Prime Intellect's reasoning LLM trained through decentralized collaboration with a 131k-token context.
Design Arena
Design Arena ranks models on real-world front-end and design tasks — websites, UI components, data viz, SVG and more — through head-to-head human votes, scored as an ELO rating.| Category | ELO | Win rate | Rank |
|---|---|---|---|
| 3D | 1164 | 45.1% | #54 |
| Website | 1159 | 42.9% | #66 |
| Code | 1157 | 42.6% | #67 |
| Game dev | 1143 | 41.1% | #66 |
| UI component | 1131 | 39.0% | #65 |
Model Description
INTELLECT-3 is a 106B-parameter Mixture-of-Experts model (12B active) post-trained from GLM-4.5-Air-Base using supervised fine-tuning (SFT) followed by large-scale reinforcement learning (RL). It offers state-of-the-art performance for its size across math,...
Author
Get the full context.
Sign up to read complete case studies, access detailed metrics, and unlock all use cases.
Author
Get the full context.
Sign up to read complete case studies, access detailed metrics, and unlock all use cases.