unsloth/Qwen3.5-0.8B-Base

Same weights, converted vs Qwen/Qwen3.5-0.8B-Base. Same weights stored in another dtype (BF16/F32 → BF16).

Tensors changed: Attention 0/216, Norms 0/129, Embeddings 0/4, MLP 0/124

1 derivatives checked

• vasidmi/Wellby-Qwen3.5-0.8B-Base: LoRA-style

On the Hub · Downloads on Model Pulse