Instructions to use Janeodum/tsaro-e2b-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Janeodum/tsaro-e2b-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Janeodum/tsaro-e2b-gguf",
	filename="1.F16-mmproj.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Janeodum/tsaro-e2b-gguf with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Janeodum/tsaro-e2b-gguf:F16
# Run inference directly in the terminal:
llama-cli -hf Janeodum/tsaro-e2b-gguf:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Janeodum/tsaro-e2b-gguf:F16
# Run inference directly in the terminal:
llama-cli -hf Janeodum/tsaro-e2b-gguf:F16

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 Janeodum/tsaro-e2b-gguf:F16
# Run inference directly in the terminal:
./llama-cli -hf Janeodum/tsaro-e2b-gguf:F16

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 Janeodum/tsaro-e2b-gguf:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Janeodum/tsaro-e2b-gguf:F16

Use Docker

docker model run hf.co/Janeodum/tsaro-e2b-gguf:F16

LM Studio
Jan

vLLM

How to use Janeodum/tsaro-e2b-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Janeodum/tsaro-e2b-gguf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Janeodum/tsaro-e2b-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Janeodum/tsaro-e2b-gguf:F16

Ollama
How to use Janeodum/tsaro-e2b-gguf with Ollama:
```
ollama run hf.co/Janeodum/tsaro-e2b-gguf:F16
```

Unsloth Studio new

How to use Janeodum/tsaro-e2b-gguf with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Janeodum/tsaro-e2b-gguf to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Janeodum/tsaro-e2b-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Janeodum/tsaro-e2b-gguf to start chatting

Pi new

How to use Janeodum/tsaro-e2b-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Janeodum/tsaro-e2b-gguf:F16

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Janeodum/tsaro-e2b-gguf:F16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Janeodum/tsaro-e2b-gguf with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Janeodum/tsaro-e2b-gguf:F16

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 Janeodum/tsaro-e2b-gguf:F16

Run Hermes

hermes

Docker Model Runner
How to use Janeodum/tsaro-e2b-gguf with Docker Model Runner:
```
docker model run hf.co/Janeodum/tsaro-e2b-gguf:F16
```

Lemonade

How to use Janeodum/tsaro-e2b-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Janeodum/tsaro-e2b-gguf:F16

Run and chat with the model

lemonade run user.tsaro-e2b-gguf-F16

List all available models

lemonade list

Tsaro Gemma 4 E2B — GGUF

Quantized GGUF build of Janeodum/tsaro-e2b, for on-device inference via llama.cpp and llama.rn.

What this model does

Tsaro is a shared safety system for Northern Nigeria. This model is its threat extraction component: it takes an unstructured report written in Hausa, Pidgin, or English and returns a structured threat signal — threat type, location, perpetrator and vehicle counts, direction of movement, time references, and a confidence score.

Model details

Quantized from: Janeodum/tsaro-e2b
Original base model: google/gemma-4-e2b-it
Quantization: Q4_K_M
Format: GGUF, for llama.cpp / llama.rn
Role in Tsaro: the E2B variant is the smaller of two on-device extraction models. It is the fallback for older or low-RAM Android devices — the Tsaro app loads the largest model the hardware can run, falling back from E4B to E2B to a hosted endpoint.

Usage

With llama.cpp:

llama-cli -m tsaro-e2b-q4_k_m.gguf -p "your threat report text here"

In a React Native app via llama.rn, the model file is bundled or downloaded on first run and loaded for offline extraction when the device has no connectivity.

Intended use and limitations

Built for community safety reporting in a specific regional context. Not a general-purpose model. Outputs are extraction assistance, not verified intelligence.