ai4burmese 1.0.0

AI4Burmese is an open Burmese-first AI ecosystem package, including Padauk and local GGUF releases for practical tool use, function calling, and accessible deployment.

AI4Burmese

AI4Burmese is an open Burmese-first AI ecosystem created by Dr. Wai Yan Nyein Naing. This Python package is the main package for AI4Burmese projects, including Padauk, a practical agentic small language model built for Burmese language understanding, tool use, function calling, and local or edge deployment.

Mission

AI4Burmese exists to help ensure Myanmar is not left behind in the age of AI.

The mission is to make AI more open, free, accessible, and useful for Burmese-speaking communities across different backgrounds, including users, students, builders, researchers, and developers.

AI4Burmese supports practical open-source AI that people can use, study, build on top of, and adapt for real local needs.

Why AI4Burmese

Burmese remains a low-resource language in AI, and practical Burmese AI tooling is still limited for many real developer and end-user workflows.

AI4Burmese focuses on useful, production-minded, and developer-friendly Burmese AI systems that are:

• open for learning and contribution
• free and accessible for broader use
• practical for real workflows
• buildable so others can extend them
• designed for Burmese-first interaction with English support

The goal is not only to release models, but also to make Burmese AI easier to use, easier to deploy, and easier to build on.

Included Ecosystem Direction

AI4Burmese is the umbrella initiative. Model families and releases can include:

• Padauk — Burmese-first agentic assistant model
• Padauk GGUF — quantized local deployment release
• Burmese-Coder — Burmese technical and coding model direction
• future open-source Burmese AI tools and runtimes

Current Model Repositories

• Source adapter model: WYNN747/ai4burmese-padauk
• GGUF local model: WYNN747/ai4burmese-padauk-gguf

End-User Install And Use

This section is the simple user flow only: install, download, run, and test the local API.

Requirements:

• Python >=3.10
• a working llama-server binary

python3 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install "ai4burmese[full]"

If your default python3 is below 3.10, use an explicit interpreter:

/opt/homebrew/bin/python3.11 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install "ai4burmese[full]"

Download Model And Run API

export AI4BURMESE_LLAMA_SERVER="$HOME/llama.cpp/build/bin/llama-server"
ai4burmese download --quant q8_0
ai4burmese serve --quant q8_0 --backend auto --host 127.0.0.1 --port 8000

Keep that terminal running while you test requests.

Test With API Call

curl http://127.0.0.1:8000/v1/models

curl http://127.0.0.1:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-no-key-required" \
  -d '{
    "model": "padauk-agent",
    "messages": [{"role": "user", "content": "မင်္ဂလာပါ"}]
  }'

Test With OpenAI Agents SDK

python -m pip install openai-agents

import asyncio

from openai import AsyncOpenAI
from agents import Agent, Runner, set_default_openai_api, set_default_openai_client, set_tracing_disabled

set_tracing_disabled(True)
set_default_openai_api("chat_completions")
set_default_openai_client(
    AsyncOpenAI(base_url="http://127.0.0.1:8000/v1", api_key="sk-no-key-required")
)

padauk = Agent(
    name="Padauk Assistant",
    model="padauk-agent",
    instructions="Reply in Burmese by default.",
)

async def main() -> None:
    result = await Runner.run(padauk, input="မင်္ဂလာပါ")
    print(result.final_output)

asyncio.run(main())

What This Package Is For

AI4Burmese is intended for:

• Burmese-first assistant applications
• local and self-hosted Burmese AI workflows
• tool-calling and function-calling assistants
• MCP-connected and skill-based systems
• practical Burmese AI experimentation
• community-accessible open AI infrastructure

Maintainer Release Validation

If you are packaging or rebuilding the model, use maintainer-facing documentation such as:

• release guide
• Hugging Face to GGUF conversion guide
• runtime compatibility notes for Gemma 4 / llama.cpp

Links

• Website: https://ai4burmese.com/
• Founder: https://www.waiyannyeinnaing.com/
• Project page: https://www.waiyannyeinnaing.com/projects/padauk
• Source repo: https://github.com/WaiYanNyeinNaing/ai4burmese
• Hugging Face source adapter: https://huggingface.co/WYNN747/ai4burmese-padauk
• Hugging Face GGUF: https://huggingface.co/WYNN747/ai4burmese-padauk-gguf

License

This package may include integrations and packaging around models released under their respective upstream licenses. Model usage remains subject to the base model license and any repository-specific release notes.

Citation

If you use AI4Burmese or Padauk in research or product work, please cite the project and model pages.

@misc{ai4burmese2026,
  title = {AI4Burmese: Open Burmese-First AI Ecosystem},
  author = {Wai Yan Nyein Naing},
  year = {2026},
  url = {https://ai4burmese.com/}
}