---
title: README
emoji: 📉
colorFrom: red
colorTo: gray
sdk: static
pinned: false
---

# SeaLLMs - Large Language Models for Southeast Asia

Welcome to the SeaLLMs project - a family of large language models tailored for Southeast Asian languages including English, Chinese, Indonesian, Vietnamese, Thai, Tagalog, Malay, Burmese, Khmer, Lao, Tamil, and Javanese. 

Unlike models primarily designed for high-resource languages like English, our mission is to democratize access to advanced language technologies for regional and potentially under-represented languages.
We aim to develop models capable of handling a variety of tasks in SEA languages, while prioritizing safety and trustworthiness within the regional context.


## SeaLLMs Models

- \[🔥NEW🔥\] SeaLLMs-v3 is here! Try the model from the [demo](https://huggingface.co/spaces/SeaLLMs/SeaLLM-Chat)!
  -  <strong style="color: red">[SeaLLMs/SeaLLMs-v3-7B-Chat](https://huggingface.co/SeaLLMs/SeaLLM3-7B-Chat)</strong>: Latest 7B chat version of SeaLLMs-v3, achieving SOTA performance of diverse tasks while specifically enhanced to be more trustworthy, exhibiting reduced hallucination and providing safe response.
  -  <strong style="color: red">[SeaLLMs/SeaLLMs-v3-1.5B-Chat](https://huggingface.co/SeaLLMs/SeaLLMs-v3-1.5B-Chat)</strong>: Latest 1.5B chat version of SeaLLMs-v3, specifically fine-tuned to follow human instructions effectively. It is designed to be resource-efficient, making it suitable for use even on your laptop.
  -  <strong style="color: red">[SeaLLMs/SeaLLMs-v3-1.5B](https://huggingface.co/SeaLLMs/SeaLLMs-v3-1.5B)</strong> and <strong style="color: red">[SeaLLMs/SeaLLMs-v3-7B](https://huggingface.co/SeaLLMs/SeaLLMs-v3-7B)</strong>: two base version models for you further conducting customized fine-tuning with your own data.
- [SeaLLMs/SeaLLM-7B-v2.5](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2.5): New SeaLLM-7B model with 7B-SOTA on many world knowledge and reasoning tasks in SEA languages.
- [SeaLLMs/SeaLLM-7B-v2](https://huggingface.co/SeaLLMs/SeaLLM-7B-v2): The most significant upgrade since SeaLLM-13B with half the size, outperforming performance across diverse multilingual tasks, from world knowledge, math reasoning, instruction following, etc.
- [SeaLLMs/SeaLLM-13B-Chat](https://huggingface.co/SeaLLMs/SeaLLM-13B-Chat): A chatbot optimized for Vietnamese 🇻🇳, Indonesian 🇮🇩, Thai 🇹🇭, Malay 🇲🇾, Khmer🇰🇭, Lao🇱🇦, Tagalog🇵🇭 and Burmese🇲🇲.

## Multilingual Evaluations for SEA
- [SeaExam](https://huggingface.co/spaces/SeaLLMs/SeaExam_leaderboard): assesses model performance using human-exam type benchmarks, reflecting the model's world knowledge and reasoning abilities.

## Quick Links
- [SeaLLM-Chatbot](https://huggingface.co/spaces/SeaLLMs/SeaLLM-Chat): online demo for the latest chatbot version of SeaLLMs
- [SeaLLMs Github Repo](https://github.com/DAMO-NLP-SG/SeaLLMs)
- [SeaLLMs Paper](https://arxiv.org/abs/2312.00738) (ACL 2024 Demo)
- \[🔥NEW🔥\] [SeaLLMs 3 Paper](https://arxiv.org/abs/2407.19672) with model training details and evaluation results!