Instructions to use rhemabible/GemmaBible with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use rhemabible/GemmaBible with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf rhemabible/GemmaBible:Q5_K_M
# Run inference directly in the terminal:
llama cli -hf rhemabible/GemmaBible:Q5_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf rhemabible/GemmaBible:Q5_K_M
# Run inference directly in the terminal:
llama cli -hf rhemabible/GemmaBible:Q5_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf rhemabible/GemmaBible:Q5_K_M
# Run inference directly in the terminal:
./llama-cli -hf rhemabible/GemmaBible:Q5_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf rhemabible/GemmaBible:Q5_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf rhemabible/GemmaBible:Q5_K_M

Use Docker

docker model run hf.co/rhemabible/GemmaBible:Q5_K_M

LM Studio
Jan

vLLM

How to use rhemabible/GemmaBible with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "rhemabible/GemmaBible"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rhemabible/GemmaBible",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/rhemabible/GemmaBible:Q5_K_M

Ollama
How to use rhemabible/GemmaBible with Ollama:
```
ollama run hf.co/rhemabible/GemmaBible:Q5_K_M
```
Unsloth Desktop

How to use rhemabible/GemmaBible with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf rhemabible/GemmaBible:Q5_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "rhemabible/GemmaBible:Q5_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use rhemabible/GemmaBible with Docker Model Runner:
```
docker model run hf.co/rhemabible/GemmaBible:Q5_K_M
```

Lemonade

How to use rhemabible/GemmaBible with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull rhemabible/GemmaBible:Q5_K_M

Run and chat with the model

lemonade run user.GemmaBible-Q5_K_M

List all available models

lemonade list

Hermes Agent

How to use rhemabible/GemmaBible with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf rhemabible/GemmaBible:Q5_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default rhemabible/GemmaBible:Q5_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use rhemabible/GemmaBible with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf rhemabible/GemmaBible:Q5_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "rhemabible/GemmaBible:Q5_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

GemmaBible

File size: 2,203 Bytes

58652ef

---
license: apache-2.0
base_model: google/gemma-4-E4B-it
tags:
  - bible
  - theology
  - gemma4
  - fine-tuned
  - qlora
language:
  - en
pipeline_tag: text-generation
---

# GemmaBible

A fine-tuned [Gemma 4 E4B](https://huggingface.co/google/gemma-4-E4B-it) model specialized in biblical scholarship, theology, and Bible study. Grounded in the Berean Standard Bible (BSB).

## What it does

- Quotes Scripture precisely from the BSB with proper citations
- Provides Greek and Hebrew word studies with transliterations and Strong's numbers
- Presents Protestant, Catholic, and Orthodox perspectives on debated topics
- Analyzes passages with scholarly hermeneutics and cited sources
- Detects and corrects common misquotations
- Stays within theological boundaries — declines non-theological requests

## Training

- **Base model**: google/gemma-4-E4B-it (4.5B effective parameters)
- **Method**: QLoRA (rank 64, alpha 64) with Unsloth
- **Data**: ~8,000 instruction examples across 11 specialized generators covering comparative theology, Greek/Hebrew exegesis, systematic theology, creedal analysis, and more
- **Hardware**: NVIDIA RTX PRO 6000 (96GB)
- **Epochs**: 3 | **Final loss**: 0.40

## Usage

### With Ollama (GGUF)

Download `merged.Q5_K_M.gguf` and create a Modelfile:

```
FROM ./merged.Q5_K_M.gguf
PARAMETER temperature 0.3
PARAMETER top_p 0.9
PARAMETER num_ctx 4096
```

```bash
ollama create gemmabible -f Modelfile
ollama run gemmabible "What does John 3:16 mean?"
```

### With Transformers

```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("rhemabible/GemmaBible")
tokenizer = AutoTokenizer.from_pretrained("rhemabible/GemmaBible")
```

## Files

| File | Format | Size | Use |
|---|---|---|---|
| `model.safetensors` | SafeTensors | ~15 GB | Full precision weights |
| `merged.Q5_K_M.gguf` | GGUF | ~5.5 GB | Ollama / LM Studio / llama.cpp |

## Limitations

- Trained on BSB text; may be less accurate with other Bible translations
- 4.5B parameters — less capacity for nuanced multi-turn theological debate compared to larger models
- Not a substitute for pastoral counsel or formal theological education