Mistral 7b Science Mcq GGUF

--- # For reference on model card metadata, see the spec: https://github.com/huggingface/hub-docs/blob/main/modelcard.md?plain=1 # Doc / guide: https://huggingface.co/docs/hub/model-cards {} --- --- # 📚 Mistral-7B Science MCQ Fine-Tuned (GGUF) ## Model Overview This repository contains a fine-tuned version of **Mistral-7B-Instruct-v0.1** optimized for **high-school level science multiple-choice question (MCQ) answering**, covering Biology, Chemistry, and Physics. The model is provided in **GGUF format** for efficient local inference (e.g., LM Studio, llama.cpp-based runtimes). --- ## 🧠 Base Model The base model used for fine-tuning: * [https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF) * Original architecture: Mistral-7B Instruct v0.1 --- ## 📊 Training Dataset The model was fine-tuned using selected subsets of the **MMLU (Massive Multitask Language Understanding)** benchmark: * High School Biology [https://huggingface.co/datasets/cais/mmlu/viewer/high_school_biology](https://huggingface.co/datasets/cais/mmlu/viewer/high_school_biology) * High School Chemistry [https://huggingface.co/datasets/cais/mmlu/viewer/high_school_chemistry](https://huggingface.co/datasets/cais/mmlu/viewer/high_school_chemistry) * High School Physics [https://huggingface.co/datasets/cais/mmlu/viewer/high_school_physics](https://huggingface.co/datasets/cais/mmlu/viewer/high_school_physics) These datasets consist of multiple-choice science questions designed to evaluate reasoning and subject knowledge at a high-school academic level. --- ## ⚙️ Model Variants ### 1. Hugging Face Fine-Tuned Model (Transformers Format) [https://huggingface.co/HangTuahMalayWarrior/mistral-7b-science-mcq-finetuned](https://huggingface.co/HangTuahMalayWarrior/mistral-7b-science-mcq-finetuned) ### 2. GGUF Model (LM Studio / llama.cpp Compatible) [https://huggingface.co/HangTuahMalayWarrior/mistral-7b-science-mcq-gguf/blob/main/mistral-7b-science-mcq-finetuned.Q4_K_M.gguf](https://huggingface.co/HangTuahMalayWarrior/mistral-7b-science-mcq-gguf/blob/main/mistral-7b-science-mcq-finetuned.Q4_K_M.gguf) Recommended quantization: **Q4_K_M** --- ## 💻 How to Use in LM Studio 1. Open **LM Studio** 2. Go to the **Models** tab 3. Search: ``` mistral-7b-science-mcq-gguf ``` 4. Download and load the model 5. Start prompting for science MCQs --- ## 🧪 Intended Use This model is optimized for: * High school science MCQ answering * Educational Q&A * Science tutoring assistance * Exam-style question practice --- ## ⚠️ Limitations * Not guaranteed to be 100% accurate in scientific reasoning *  * May hallucinate or guess answers when uncertain * Not a replacement for verified textbooks or academic sources * Performance depends on prompt quality and context --- ## 📦 Why GGUF Version? The GGUF version is provided for: * Fast CPU/GPU inference * Local deployment (offline usage) * Compatibility with LM Studio and llama.cpp-based tools * Lower memory requirements via quantization --- ## 📌 Notes We also uploaded this model for **open archival and public access**, allowing anyone to download and experiment with the fine-tuned science MCQ model. --- ## 🙌 Acknowledgements * Mistral AI for the base model * TheBloke * CAIS MMLU dataset contributors ---