openbmb/MiniCPM-V-4-gguf

Quantization (fewer layers) vs openbmb/MiniCPM-V-4. openbmb/MiniCPM-V-4-gguf matches part of openbmb/MiniCPM-V-4 (291 of its 741 tensors; the rest of openbmb/MiniCPM-V-4 has no counterpart). On the shared part: openbmb/MiniCPM-V-4-gguf is a GGUF quantization (Q8_0) of the same weights: 65 tensors match exactly and 226 within quantization error.

Tensors changed: Embeddings 0/1, Norms 0/65, MLP 0/96, Attention 0/128, lm_head 0/1

On the Hub · Downloads on Model Pulse