blascotobasco/GLM-4.6V-96E
LoRA-style vs zai-org/GLM-4.6V. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, up_proj changed, in 45 of 46 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/4, Norms 0/145, MLP 1582/1841, Attention 0/370, lm_head 0/1, other 0/3
2 derivatives checked
- mradermacher/GLM-4.6V-96E-i1-GGUF: Quantization
- mradermacher/GLM-4.6V-96E-GGUF: Quantization