extraltodeus/Qwen3.5-9B-Nikusui-v1
Partial fine-tune vs Qwen/Qwen3.5-9B-Base. Partial change: 55% of the weights changed (attention, lm_head, mlp), in 29 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 0/5, MLP 58/211, Attention 108/336, other 0/1, Norms 0/221
3 derivatives checked
- mradermacher/Qwen3.5-9B-Nikusui-v1-i1-GGUF: Quantization
- mradermacher/Qwen3.5-9B-Nikusui-v1-GGUF: Same weights, converted
- llmfan46/Qwen3.5-9B-Nikusui-v1-Uncensored-Heretic-Native-MTP-Preserved: Abliterated