开源泰语大模型OpenThaiGPT 1.5，基于Qwen v2.5微调，支持多轮对话与工具调用。
OpenThaiGPT 1.5: A Thai-Centric Open Source Large Language Model
- 在200万条泰语指令数据上微调Qwen v2.5，构建专注泰语的对话模型
- 在多项泰语任务中超越现有开源模型，性能达顶尖水平
- 支持多轮对话、RAG和工具调用，适合本地部署与实际应用
OpenThaiGPT 1.5 是基于 Qwen v2.5 构建的先进泰语聊天模型，在超过 200 万条泰语指令对上进行微调。本文从工程角度阐述模型的开发、能力与性能表现，涵盖架构设计、训练流程及关键特性，包括多轮对话支持、与检索增强生成（RAG）兼容、工具调用功能。基准测试显示，OpenThaiGPT 1.5 在多项泰语任务中表现领先于其他开源模型。同时讨论了 GPU 显存需求与部署策略等实际问题。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。