nobrand/KULLM-R

Partial fine-tune vs Qwen/Qwen3-8B. Partial change: 84% of the weights changed (attention, mlp, norms), in 36 of 36 layers.

Tensors changed: Embeddings 0/1, Norms 56/145, MLP 108/108, Attention 141/144, lm_head 0/1

2 derivatives checked

• mradermacher/KULLM-R-i1-GGUF: Quantization
• mradermacher/KULLM-R-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse