mradermacher/Gemma3-Rhino-27B-RAG-GGUF

Same weights, converted (fewer layers) vs QomSSLab/Gemma3-Rhino-27B-RAG. mradermacher/Gemma3-Rhino-27B-RAG-GGUF matches part of QomSSLab/Gemma3-Rhino-27B-RAG (444 of its 883 tensors; the rest of QomSSLab/Gemma3-Rhino-27B-RAG has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/205, MLP 0/102, Attention 0/136

On the Hub · Downloads on Model Pulse