CoolFace
Modelpublic

build-small-hackathon/MiniCPM4-8B-PaperProf-GGUF

sourceHugging Faceapache-2.0updated 3mo agoView on Hugging Face
0likes30downloads
Model Card

MiniCPM4-8B-PaperProf-GGUF

GGUF quantizations of build-small-hackathon/MiniCPM4-8B-PaperProf, the fine-tuned exam-question generator behind PaperProf.

FileQuantSizeUse
minicpm4-8b-paperprof-Q4_K_M.ggufQ4KM~4.9 GBrecommended, used by the Space
minicpm4-8b-paperprof-f16.ggufF16~16 GBfull precision reference

Usage with llama.cpp

bash
llama-cli -hf build-small-hackathon/MiniCPM4-8B-PaperProf-GGUF -p "your prompt"

Usage with llama-cpp-python

python
from llama_cpp import Llama
llm = Llama.from_pretrained("build-small-hackathon/MiniCPM4-8B-PaperProf-GGUF", filename="*Q4_K_M.gguf", n_gpu_layers=-1)

Built for the Build Small Hackathon, June 2026, by Team PaperProf (EPITA).