steampunque/MiniCPM-V-4_5-MP-GGUF

Quantization (fewer layers) vs openbmb/MiniCPM-V-4_5. steampunque/MiniCPM-V-4_5-MP-GGUF matches part of openbmb/MiniCPM-V-4_5 (399 of its 849 tensors; the rest of openbmb/MiniCPM-V-4_5 has no counterpart). On the shared part: steampunque/MiniCPM-V-4_5-MP-GGUF is a GGUF quantization (Q6_K_H) of the same weights: 145 tensors match exactly and 254 within quantization error.

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

On the Hub · Downloads on Model Pulse