LingJianAI/Lingjian-1.5

Full fine-tune vs Qwen/Qwen3-VL-8B-Instruct. Fine-tune: 93% of the weights changed, across 36 of 36 layers.

Tensors changed: Embeddings 4/4, Norms 238/261, MLP 232/232, Attention 252/252, lm_head 0/1

1 derivatives checked

• mradermacher/Lingjian-1.5-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse