The general-purpose workhorse of the OpenThai family — 98.2% Thai language accuracy, strongest of the family on Thai coding (LiveCodeBench-TH 32.43). These are the official quantizations, built from the source weights by the OpenThai team.

For step-by-step reasoning use R1 32B — smaller and better at it. For a laptop, use 1.5 7B.

Quants

File	Quant	Size	Notes
`openthaigpt-1.6-72b-instruct.Q4_K_M.gguf`	Q4_K_M	~44 GB	Recommended. 2× 24 GB GPUs or a 64 GB Mac.
`openthaigpt-1.6-72b-instruct.Q5_K_M.gguf`	Q5_K_M	~51 GB	Higher quality.
`openthaigpt-1.6-72b-instruct.Q8_0.gguf`	Q8_0	~77 GB	Near-lossless.

Usage

Ollama

ollama run hf.co/openthaigpt/openthaigpt-1.6-72b-instruct-GGUF:Q4_K_M

llama.cpp

bash

llama-cli -m openthaigpt-1.6-72b-instruct.Q4_K_M.gguf \
  -p "เขียนฟังก์ชัน Python แปลงเลขไทยเป็นเลขอารบิก พร้อมอธิบายโค้ด" -n 2048 --temp 0.7

Uses the ChatML template (<|im_start|> / <|im_end|>), embedded in the GGUF. Recommended sampling: temperature=0.7, top_p=0.9.

Citation

bibtex

@misc{yuenyong2025openthaigpt16r1thaicentric,
      title={OpenThaiGPT 1.6 and R1: Thai-Centric Open Source and Reasoning Large Language Models},
      author={Sumeth Yuenyong and Thodsaporn Chay-intr and Kobkrit Viriyayudhakorn},
      year={2025},
      eprint={2504.01789},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2504.01789}
}

OpenThai (formerly OpenThaiGPT) — free, open-weight Thai large language models from AIEAT and iApp Technology. With thanks to the community members who published unofficial GGUF conversions before these existed.