Instructions to use Roopalgn/sahayak-tb-gemma4-e2b-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Roopalgn/sahayak-tb-gemma4-e2b-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Roopalgn/sahayak-tb-gemma4-e2b-gguf",
	filename="gemma-4-e2b-it.F16-mmproj.gguf",
)

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

llama-cpp-python

How to use Roopalgn/sahayak-tb-gemma4-e2b-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Roopalgn/sahayak-tb-gemma4-e2b-gguf",
	filename="gemma-4-e2b-it.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 Roopalgn/sahayak-tb-gemma4-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 Roopalgn/sahayak-tb-gemma4-e2b-gguf:F16
# Run inference directly in the terminal:
llama-cli -hf Roopalgn/sahayak-tb-gemma4-e2b-gguf:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Roopalgn/sahayak-tb-gemma4-e2b-gguf:F16
# Run inference directly in the terminal:
llama-cli -hf Roopalgn/sahayak-tb-gemma4-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 Roopalgn/sahayak-tb-gemma4-e2b-gguf:F16
# Run inference directly in the terminal:
./llama-cli -hf Roopalgn/sahayak-tb-gemma4-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 Roopalgn/sahayak-tb-gemma4-e2b-gguf:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Roopalgn/sahayak-tb-gemma4-e2b-gguf:F16

Use Docker

docker model run hf.co/Roopalgn/sahayak-tb-gemma4-e2b-gguf:F16

LM Studio
Jan

vLLM

How to use Roopalgn/sahayak-tb-gemma4-e2b-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Roopalgn/sahayak-tb-gemma4-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": "Roopalgn/sahayak-tb-gemma4-e2b-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Roopalgn/sahayak-tb-gemma4-e2b-gguf:F16

Ollama
How to use Roopalgn/sahayak-tb-gemma4-e2b-gguf with Ollama:
```
ollama run hf.co/Roopalgn/sahayak-tb-gemma4-e2b-gguf:F16
```

Unsloth Studio new

How to use Roopalgn/sahayak-tb-gemma4-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 Roopalgn/sahayak-tb-gemma4-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 Roopalgn/sahayak-tb-gemma4-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 Roopalgn/sahayak-tb-gemma4-e2b-gguf to start chatting

Docker Model Runner
How to use Roopalgn/sahayak-tb-gemma4-e2b-gguf with Docker Model Runner:
```
docker model run hf.co/Roopalgn/sahayak-tb-gemma4-e2b-gguf:F16
```

Lemonade

How to use Roopalgn/sahayak-tb-gemma4-e2b-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Roopalgn/sahayak-tb-gemma4-e2b-gguf:F16

Run and chat with the model

lemonade run user.sahayak-tb-gemma4-e2b-gguf-F16

List all available models

lemonade list

SAHAYAK-TB — Fine-tuned Gemma 4 E2B for TB Dropout Risk Assessment

What This Model Does

Takes natural-language multi-visit notes from a DOTS (Directly Observed Treatment) provider and outputs a structured JSON risk assessment for TB treatment adherence dropout.

Input: 2-4 visit notes in English, Hindi, or mixed (as DOTS providers actually write)

Output: JSON with dropout risk tier (low/moderate/high/critical), evidence-cited risk factors with published odds ratios and PMIDs, and NTEP protocol action recommendations.

Key Specs


Base model	Gemma 4 E2B-it
Fine-tuning	QLoRA (r=16, alpha=16, all-linear) via Unsloth
Training data	500 synthetic cases from 15 published TB adherence risk factors
Quantization	Q4_K_M (3.43 GB) for CPU inference
Accuracy	90% tier accuracy with post-processing pipeline
Schema compliance	100% (constrained JSON output)
Inference	CPU-only via llama-cpp-python, 24-88 seconds

Usage with llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
    repo_id="Roopalgn/sahayak-tb-gemma4-e2b-gguf",
    filename="gemma-4-e2b-it.Q4_K_M.gguf",
    n_ctx=4096,
    n_gpu_layers=0,
)

response = llm.create_chat_completion(
    messages=[
        {"role": "system", "content": SYSTEM_PROMPT},
        {"role": "user", "content": visit_notes},
    ],
    response_format={"type": "json_object", "schema": RISK_SCHEMA},
    temperature=0.0,
)

See full system prompt and schema at: github.com/Roopalgn/gemma4good

Live Demo

huggingface.co/spaces/Roopalgn/sahayak-tb-demo

Training Details

Platform: Kaggle T4 GPU (free tier, $0 cost)
Training time: 33 minutes
Loss: 3.04 to 0.0115
Method: SFT with train_on_responses_only

Data Sources

15 risk factors with published odds ratios from Indian TB cohort studies:

Santha et al. 2002 (PMID 12234133)
Vijay et al. 2010 (PMID 20386611)
Kumar et al. 2024 (PMID 39067959)
WHO Treatment of TB Guidelines (4th ed.)

No patient-identifiable data used. Fully synthetic training data.

License

CC-BY 4.0

Built for

Gemma 4 Good Hackathon — Health & Sciences track + llama.cpp Prize

Downloads last month: 943

GGUF

Model size

5B params

Architecture

gemma4

Hardware compatibility

4-bit

16-bit

Roopalgn
/

sahayak-tb-gemma4-e2b-gguf