Live page · Day archive

Dory1.0-4B: A Looped Normalized-GPT

Read the original at arXiv

Summary

Dory1.0-4B is a 4-billion-parameter looped language model trained on 6 trillion tokens that outperforms larger models like Qwen3.5-9B and Gemma 4 12B on 12 benchmarks.

Carried by: arXiv, HF Daily Papers. First seen: .