Instructions to use axom-ai/AxomAI-Sarkar with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use axom-ai/AxomAI-Sarkar 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 axom-ai/AxomAI-Sarkar:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf axom-ai/AxomAI-Sarkar:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf axom-ai/AxomAI-Sarkar:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf axom-ai/AxomAI-Sarkar:Q4_K_M

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf axom-ai/AxomAI-Sarkar:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf axom-ai/AxomAI-Sarkar:Q4_K_M

Use Docker

docker model run hf.co/axom-ai/AxomAI-Sarkar:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use axom-ai/AxomAI-Sarkar with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/axom-ai/AxomAI-Sarkar:Q4_K_M

Ollama
How to use axom-ai/AxomAI-Sarkar with Ollama:
```
ollama run hf.co/axom-ai/AxomAI-Sarkar:Q4_K_M
```
Unsloth Desktop

How to use axom-ai/AxomAI-Sarkar with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf axom-ai/AxomAI-Sarkar: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": "axom-ai/AxomAI-Sarkar:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use axom-ai/AxomAI-Sarkar with Docker Model Runner:
```
docker model run hf.co/axom-ai/AxomAI-Sarkar:Q4_K_M
```

Lemonade

How to use axom-ai/AxomAI-Sarkar with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull axom-ai/AxomAI-Sarkar:Q4_K_M

Run and chat with the model

lemonade run user.AxomAI-Sarkar-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use axom-ai/AxomAI-Sarkar with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf axom-ai/AxomAI-Sarkar: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 axom-ai/AxomAI-Sarkar:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use axom-ai/AxomAI-Sarkar with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf axom-ai/AxomAI-Sarkar: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 "axom-ai/AxomAI-Sarkar: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"

You need to agree to share your Hugging Face information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

🏛️ AxomAI Sarkar (অসম চৰকাৰ AI সহায়ক)

The Sovereign 492M Small Language Model (SLM) & Citizen Knowledge Base for Government of Assam

AxomAI Sarkar is a specialized, lightweight, high-speed regional language model trained on 3.3 Million+ verified citizen instruction pairs (4.81 GB) and 21 master administrative datasets. Engineered for 100% CPU inference (zero GPU required), it runs smoothly on standard laptops, desktops, and edge kiosks.

⚡ Quick Model Specifications

Specification	Value / Metric
Base Architecture	Qwen 2.5 Base (492M parameters)
Hardware Requirement	100% CPU Native (Zero GPU required)
Context Window (`n_ctx`)	4,096 tokens
Inference Speed	40 – 75 tokens/sec on standard quad-core CPU
Languages Supported	অসমীয়া (Assamese), English, हिन्दी / Hinglish
License	Apache-2.0

📊 Dataset Summary (3.3M+ Instructions • 21 Master Datasets)

Dataset Layer	Total Volume	Scope & Details
Instruction Training Mega-Corpus	3,300,227 Rows (4.81 GB)	Verified Q&A pairs covering citizen workflows, disputes, appeals, and post-2023 Delimitation (130 LACs).
Master Reference Datasets	21 Datasets (JSON, CSV, JSONL)	Deeply structured data across 21 core public sectors with official rules, fees, timelines, and links.
Knowledge Base Modules	48 Modules (482 Records)	Modular profiles of all 35 Districts, Autonomous Councils (BTR, KAAC, DHAC), and state directories.

Core Administrative Areas Covered:

Land Revenue & Mission Basundhara: Basundhara (1.0, 2.0, 3.0), AP to PP conversion, inheritance namjari, e-Khajana, Dharitree Jamabandi.
Citizen Certificates (Sewa Setu / ARTPS): PRC, Caste (SC/ST/OBC/MOBC), NCL, Income, Next of Kin, Bakijai, Trade License.
Flagship Schemes: Orunodoi 3.0, Nijut Moina, CMAAA (Rs 2L–5L grant), Ayushman Asom (MMJAY), Pragyan Bharati Scooty.
Public Safety & Police: Regular & Zero FIR, 1930 Cybercrime, PCC, Tenant verification, ERSS 112, Anti-Corruption 1064 traps.
Public Utilities & Transport: APDCL billing & smart meters (1912), PM Surya Ghar solar subsidy, Faceless LL, DL, RC, e-Challan.
Disaster & Grievance Redressal: ASDMA flood relief (Rs 4L ex-gratia), SDRF norms, CPGRAMS, ARTPS appeals (Rs 250/day penalty), Banking & Insurance Ombudsmen Guwahati.

🚀 Quick Start Guide

1. Run with Ollama

ollama run hf.co/axom-ai/AxomAI-Sarkar:AxomAI-Sarkar-0.5B-Q4_K_M.gguf

2. Run with Python (`llama-cpp-python` — 100% CPU)

from llama_cpp import Llama
from huggingface_hub import hf_hub_download

# Download GGUF binary
model_path = hf_hub_download(
    repo_id="axom-ai/AxomAI-Sarkar",
    filename="AxomAI-Sarkar-0.5B-Q4_K_M.gguf"
)

# Initialize on standard CPU (4 threads, 4096 context)
llm = Llama(model_path=model_path, n_ctx=4096, n_threads=4)

# Query model using ChatML prompt format
prompt = "<|im_start|>user\nMission Basundhara 3.0 mein AP to PP patta conversion ke rules kya hain?<|im_end|>\n<|im_start|>assistant\n"
output = llm(prompt, max_tokens=300, temperature=0.2)
print(output["choices"][0]["text"])

🏛️ Organization & License

Organization: AxomAI
License: Apache-2.0

Downloads last month: -

GGUF

Model size

0.6B params

Architecture

qwen2

Hardware compatibility

4-bit

Model tree for axom-ai/AxomAI-Sarkar

Base model

Qwen/Qwen2.5-0.5B

Finetuned

Qwen/Qwen2.5-0.5B-Instruct

Quantized

(373)

this model

You need to agree to share your Hugging Face information to access this model

🏛️ AxomAI Sarkar (অসম চৰকাৰ AI সহায়ক)

⚡ Quick Model Specifications

📊 Dataset Summary (3.3M+ Instructions • 21 Master Datasets)

Core Administrative Areas Covered:

🚀 Quick Start Guide

1. Run with Ollama

2. Run with Python (llama-cpp-python — 100% CPU)

🏛️ Organization & License

Model tree for axom-ai/AxomAI-Sarkar

2. Run with Python (`llama-cpp-python` — 100% CPU)