← Back to feed

Nenya, a 0.5B-active base trained on 60B tokens, rivals Qwen3.5-2B

AI · Studies · 2026-11-10 00:01

Hugging Face has released Nenya, a base model with only 0.5 billion active parameters trained on 60 billion tokens. Despite its compact size, Nenya delivers performance comparable to the larger Qwen3.5-2B, highlighting significant efficiency gains in small-scale language modeling.

Source: huggingface.co


Auto-generated summary for reference only. RSS