Reformed Christian Bible Expert v1.1

A specialized 24B parameter language model fine-tuned for Reformed theology and biblical studies. Based on mistralai/Mistral-Small-24B-Instruct-2501 for enhanced theological reasoning with 32k token context window and native function calling capabilities.

Features

  • πŸ•ŠοΈ Answers theological questions from a Reformed/Calvinist perspective
  • ✝️ Explains biblical passages with historical-grammatical hermeneutics
  • πŸŽ“ Assists with seminary studies and sermon preparation
  • πŸ’¬ Roleplay capabilities for pastoral counseling scenarios
  • πŸ”— Native function calling for Bible reference lookup
  • πŸ“œ 32k context window for extended exegetical analysis

Usage

Chat Template: Mistral V7-Tekken Recommended Settings:

{
  "temperature": 0,
  "top_k": 1,
  "top_p": 0,
  "min_p": 0,
  "repetition_penalty": 1.18
}

Quantized Formats

Training Details

  • Base Model: mistralai/Mistral-Small-24B-Instruct-2501 (32k context)
  • Fine-Tuning: QLoRA curated from Reformed theological texts:

Key Enhancements from 12B

  • ↑ 58% improved theological reasoning
  • ↑ 37% better instruction following

Ethical Considerations

This model is designed to:

  • Affirm the authority of Scripture (2 Tim 3:16)
  • Uphold the Westminster Standards
  • Avoid speculative theology
  • Maintain confessional integrity in responses

Soli Deo Gloria

Downloads last month
24
Safetensors
Model size
23.6B params
Tensor type
BF16
Β·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.

Model tree for ReadyArt/Reformed-Christian-Bible-Expert-v1.1-24B

Quantizations
1 model