An instruction-following dataset of 1,260 Burmese-language microbiology Q&A pairs — ideal for fine-tuning Burmese LLMs on scientific knowledge. Covers bacteria, viruses, fungi, and core microbiology concepts.
A 1,050-row instruction dataset in the Burmese language covering Myanmar agriculture — crops, soil, farming techniques, and rural knowledge. Designed for domain-specific fine-tuning on agricultural topics relevant to Myanmar.
A Burmese-language dataset on Mpox (monkeypox) covering symptoms, transmission, prevention, and treatment — curated for medical NLP and health information systems in Myanmar. Timely and relevant to public health needs.
A 1,920-row Burmese-language roleplay and conversational dataset for instruction-following and persona-based fine-tuning. Enables natural Burmese dialogue in chatbots, assistants, and interactive AI applications.
A 2,040-row Burmese instruction dataset derived from Myanmar national tuberculosis clinical guidelines. The largest dataset in this collection — essential for medical AI, clinical decision support, and Burmese health chatbots.
A 105-row Burmese-language football (soccer) knowledge dataset covering rules, positions, formations, leagues, tournaments, and terminology — designed for NLP research and LLM development in Myanmar language. Includes Q&A pairs on offside, penalties, cards, player roles, and major competitions like FIFA World Cup and Premier League.
A 221-row Rakhine/Arakan language proverbs dataset — a low-resource language spoken in Rakhine State, Myanmar. Proverbs are extracted and summarized from "ဥပမာစုံ၊ ရခိုင်စကားပုံ။" published in 1996. Essential for preserving and researching the Rakhine language in NLP and LLM development.
Published a Burmese language dataset? Use #MyanmarGPTMovement on Hugging Face and submit a PR to get listed here.