Instructions to use vikasit-ai/Vikasit-AI-0.8B-Writer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use vikasit-ai/Vikasit-AI-0.8B-Writer with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="vikasit-ai/Vikasit-AI-0.8B-Writer",
	filename="Vikasit-AI-Writer-0.8B-IQ4_XS-Final.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 vikasit-ai/Vikasit-AI-0.8B-Writer with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS
# Run inference directly in the terminal:
llama-cli -hf vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS
# Run inference directly in the terminal:
llama-cli -hf vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS

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 vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS
# Run inference directly in the terminal:
./llama-cli -hf vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS

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 vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS
# Run inference directly in the terminal:
./build/bin/llama-cli -hf vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS

Use Docker

docker model run hf.co/vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS

LM Studio
Jan

vLLM

How to use vikasit-ai/Vikasit-AI-0.8B-Writer with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "vikasit-ai/Vikasit-AI-0.8B-Writer"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "vikasit-ai/Vikasit-AI-0.8B-Writer",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS

Ollama
How to use vikasit-ai/Vikasit-AI-0.8B-Writer with Ollama:
```
ollama run hf.co/vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS
```

Unsloth Studio new

How to use vikasit-ai/Vikasit-AI-0.8B-Writer 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 vikasit-ai/Vikasit-AI-0.8B-Writer 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 vikasit-ai/Vikasit-AI-0.8B-Writer to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for vikasit-ai/Vikasit-AI-0.8B-Writer to start chatting

Docker Model Runner
How to use vikasit-ai/Vikasit-AI-0.8B-Writer with Docker Model Runner:
```
docker model run hf.co/vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS
```

Lemonade

How to use vikasit-ai/Vikasit-AI-0.8B-Writer with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull vikasit-ai/Vikasit-AI-0.8B-Writer:IQ4_XS

Run and chat with the model

lemonade run user.Vikasit-AI-0.8B-Writer-IQ4_XS

List all available models

lemonade list

🇮🇳 Vikasit AI Writer 0.8B (IQ4_XS)

Vikasit AI Writer 0.8B is a next-generation, sovereign Indian language model developed by Chandorkar Technologies. It is built on the Qwen 3.5 Hybrid Gated DeltaNet architecture, specifically optimized to provide high-density intelligence in a compact, on-device footprint.

As part of the Vikasit AI mission, this model is designed to provide India with a high-performance, locally-executable AI stack that respects data sovereignty and cultural context.

🚀 Model Highlights

Architecture: Hybrid Gated DeltaNet (O(1) memory for linear attention).
Parameters: 0.8 Billion.
Context Window: 256,144 tokens (Native).
Quantization: Custom iMatrix IQ4_XS (GGUF).
Size: ~504 MB.
Identity: Native Vikasit AI persona, refined for Indian professional and cultural nuances.

🛠 Quick Start (Ollama)

You can run this model locally with a single command:

ollama run vikasit-ai/writer:0.8b

Downloads last month: 534

GGUF

Model size

0.8B params

Architecture

qwen35

Hardware compatibility

4-bit