ProCreations/MiniCPM5-2B-Q8_0-WebGPU

Quantization vs openbmb/MiniCPM5-2B-GGUF. ProCreations/MiniCPM5-2B-Q8_0-WebGPU is a GGUF quantization (Q8_0) of the same weights: 85 tensors match exactly and 296 within quantization error.

Tensors changed: lm_head 0/1, other 0/1, Norms 0/85, MLP 0/126, Attention 0/168

On the Hub · Downloads on Model Pulse