LightOriginsHQ/LightNav-0

Full fine-tune vs Qwen/Qwen3-VL-4B-Instruct. Fine-tune: 100% of the weights changed, across 36 of 36 layers.

Tensors changed: Norms 249/249, MLP 220/220, Attention 240/240, Embeddings 3/3

1 derivatives checked

• prithivMLmods/LightNav-0-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse