OpenThaiGPT 1.5: A Thai-Centric Open Source Large Language Model

TL;DR

OpenThaiGPT 1.5 is a state-of-the-art Thai language chat model based on Qwen v2.5, finetuned on extensive instruction data, supporting multi-turn conversations, RAG, and tool-calling, with strong benchmark performance.

Contribution

This paper introduces OpenThaiGPT 1.5, a new open-source Thai language model with advanced features and improved performance over existing models.

Findings

Outperforms other open-source Thai models on benchmarks

Supports multi-turn conversations and tool integration

Demonstrates practical deployment strategies

Abstract

OpenThaiGPT 1.5 is an advanced Thai language chat model based on Qwen v2.5, finetuned on over 2,000,000 Thai instruction pairs. This report provides an engineering perspective on the model's development, capabilities, and performance. We discuss the model's architecture, training process, and key features, including multi-turn conversation support, Retrieval Augmented Generation (RAG) compatibility, and tool-calling functionality. Benchmark results demonstrate OpenThaiGPT 1.5's state-of-the-art performance on various Thai language tasks, outperforming other open-source Thai language models. We also address practical considerations such as GPU memory requirements and deployment strategies.