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

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use prawinin/vidhi 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 prawinin/vidhi:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf prawinin/vidhi:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf prawinin/vidhi:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf prawinin/vidhi:Q4_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 prawinin/vidhi:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf prawinin/vidhi:Q4_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 prawinin/vidhi:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf prawinin/vidhi:Q4_K_M

Use Docker

docker model run hf.co/prawinin/vidhi:Q4_K_M

LM Studio
Jan

vLLM

How to use prawinin/vidhi with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/prawinin/vidhi:Q4_K_M

Ollama
How to use prawinin/vidhi with Ollama:
```
ollama run hf.co/prawinin/vidhi:Q4_K_M
```
Unsloth Desktop

How to use prawinin/vidhi with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf prawinin/vidhi:Q4_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": "prawinin/vidhi:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use prawinin/vidhi with Docker Model Runner:
```
docker model run hf.co/prawinin/vidhi:Q4_K_M
```

Lemonade

How to use prawinin/vidhi with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull prawinin/vidhi:Q4_K_M

Run and chat with the model

lemonade run user.vidhi-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use prawinin/vidhi with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf prawinin/vidhi:Q4_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 prawinin/vidhi:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use prawinin/vidhi with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf prawinin/vidhi:Q4_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 "prawinin/vidhi:Q4_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"

Vidhi

Vidhi is an Indian legal language model fine-tuned on a corpus of Supreme Court of India judgments and Central Government legislation. It is designed to assist with legal research, statutory interpretation, and understanding of Indian jurisprudence across criminal, civil, constitutional, and commercial law.

The model runs entirely on consumer hardware. No API, no internet connection, no subscription.

Quick Start

Ollama (recommended)

ollama run hf.co/prawinin/vidhi

Requires Ollama installed. The model loads in under 10 seconds on a 4 GB GPU and under 20 seconds on CPU.

llama.cpp

llama-cli -hf prawinin/vidhi

What It Knows

Vidhi was trained on over 100,000 Indian legal documents spanning:

Supreme Court judgments across constitutional, criminal, civil, and commercial benches
High Court orders and doctrine from multiple jurisdictions
Central Acts and statutory provisions covering major legislation in force in India

The model has working knowledge of the Indian Constitution, the Indian Penal Code, the Code of Criminal Procedure, the Code of Civil Procedure, the Negotiable Instruments Act, the Consumer Protection Act, the Transfer of Property Act, the Companies Act, GST legislation, and several hundred other Central Acts.

Model Details

Property	Value
Base model	Llama 3.2 3B Instruct
Fine-tuning method	LoRA (rank 16, alpha 16)
Context window	8,192 tokens
Quantization	Q4_K_M
File size	~2.0 GB
Minimum VRAM	3 GB (GPU) or 8 GB RAM (CPU)

Limitations

Vidhi is a research and educational tool. It is not a substitute for qualified legal advice from a licensed advocate.
The model has a knowledge cutoff and may not reflect recent amendments, new legislation, or judgments delivered after its training data was collected.
Outputs should always be independently verified against authoritative sources before being relied upon in any professional context.

License

This model is made available for personal, non-commercial evaluation only.

You may not redistribute, sublicense, sell, modify, or use this model or any derivative of it in any product, service, or research publication without explicit written permission.

For commercial licensing, enterprise deployment, partnerships, or any other inquiries: prawin@vyapai.tech

Built with Unsloth. Underlying base model subject to the Meta Llama 3.2 Community License.

Downloads last month: 261

GGUF

Model size

3B params

Architecture

llama

Hardware compatibility

4-bit

Model tree for prawinin/vidhi

Base model

meta-llama/Llama-3.2-3B-Instruct

Quantized

(534)

this model