Rombo-Org/Rombo-LLM-V3.1-QWQ-32b
Full fine-tune vs Qwen/QwQ-32B. Fine-tune: 95% of the weights changed, across 64 of 64 layers.
Tensors changed: Embeddings 0/1, Norms 57/129, MLP 187/192, Attention 403/448, lm_head 1/1
5 derivatives checked
- bartowski/Rombo-Org_Rombo-LLM-V3.1-QWQ-32b-GGUF: Quantization
- BasedAGI/Rombo-LLM-V3.1-QWQ-32B-i1-GGUF: Quantization
- mradermacher/Rombo-LLM-V3.1-QWQ-32b-i1-GGUF: Quantization
- DevQuasar-10/Rombo-Org.Rombo-LLM-V3.1-QWQ-32b-GGUF: Same weights, converted
- mradermacher/Rombo-LLM-V3.1-QWQ-32b-GGUF: Quantization