SynastriaNetworks/Uirapuru-U1-9B

Full fine-tune vs Qwen/Qwen3.5-9B-Base. Fine-tune: 96% of the weights changed, across 32 of 32 layers.

Tensors changed: lm_head 1/1, Embeddings 3/5, MLP 103/211, Attention 201/336, other 1/1, Norms 113/221

3 derivatives checked

• SynastriaNetworks/U1-0817: LoRA-style
• mradermacher/Uirapuru-U1-9B-GGUF: Same weights, converted
• mradermacher/Uirapuru-U1-9B-i1-GGUF: Quantization

On the Hub · Downloads on Model Pulse