zenlm/zen-nano-0.6b

Full fine-tune vs Qwen/Qwen3-0.6B. Fine-tune: 100% of the weights changed, across 28 of 28 layers.

Tensors changed: Embeddings 1/1, Norms 61/113, MLP 84/84, Attention 112/112

2 derivatives checked

• mradermacher/zen-nano-0.6b-i1-GGUF: Full fine-tune
• mradermacher/zen-nano-0.6b-GGUF: Full fine-tune

On the Hub · Downloads on Model Pulse