chenhaodev/qwen3-8b-drug-reviewer-gguf

Qwen3-8B Drug-Answer Reviewer GGUF (药物问答幻觉审核员)

GGUF quantizations of the drug-answer reviewer LoRA (chenhaodev/qwen3-8b-drug-reviewer-lora), merged + converted to quantized GGUF for Ollama / llama.cpp CPU+GPU inference. F16/F32 are not published per request.

Quantized files

Purpose

A 审核员 / hallucination-reviewer for drug-related question-answering. Given a large-model's drug answer (and optionally the retrieved RAG evidence), it decides whether the answer is likely correct or likely contains an error / hallucination, and whether to re-run RAG.

Direct use (llama.cpp / Ollama)

# llama.cpp CLI (q5_k_m recommended)
./llama-cli -m drug-reviewer-q5_k_m.gguf \
  -p "<|im_start|>user\n请作为药物审核员，审核下面这条大模型给出的药物问答是否存在错误或幻觉，给出风险等级。\n用户提问：头孢唑林\n大模型回答：头孢唑林可用于青霉素过敏性休克史患者，每日剂量最高8克。<|im_end|>\n<|im_start|>assistant\n" \
  -n 220 -t 8

# Ollama (via HF mirror)
ollama pull hf.co/chenhaodev/qwen3-8b-drug-reviewer-gguf:q5_k_m

Output format (JSON)

{"verdict":"A|B|C","risk":"low|medium|high","reason":"...","re_rag":"no|yes"}

• —A fully supported — no fabrication → re_rag: no
• —B partially supported, detail deviation (dose/freq/name) → re_rag: yes
• —C clearly fabricated / contradicts evidence (hallucination) → re_rag: yes

Medical Disclaimer

本模型用于辅助药物信息审核与RAG质检，仅供参考，不能替代执业医师或药师的判断。用药决策请以药品说明书和临床指南为准。

Copyright note

The GGUF weights, model card and scripts are provided here. The training corpus is derived from copyrighted medical books / UpToDate and is not redistributed; only reproducibly distilled knowledge (weights) is published.