← sources

HN: 0.5B base model trained on 60B tokens rivals Qwen3.5-2B trained on 30T+ tokens

source: hackernewspublished 2026-11-13 23:40ingested 2026-11-15 02:47

retained text

Points: 12 | Comments: 2
Author: AZDR
Link: https://huggingface.co/ArsalanZ/Nenya-2.5B-A0.5B-Base
0.5B base model trained on 60B tokens rivals Qwen3.5-2B trained on 30T+ tokens

document fetched 2026-11-20 02:51