**腾讯推出混元文生图大模型，开源训练代码及新插件**
近日，腾讯公司宣布了其全新的自然语言处理创新成果——混元文生图大模型的开源训练代码。与此同时，备受瞩目的LoRA与ControlNet两大插件也同步对外开源。这项技术的推出标志着中文自然语言处理领域迈入了新的里程碑。
据了解，混元文生图大模型（简称“混元DiT模型”）此次全面开源的训练代码，为用户提供了更为便捷的使用体验。作为纯中文原生模型，该模型在训练过程中可直接使用中文数据与标签，无需繁琐的数据翻译过程。这一创新不仅简化了操作流程，还大大提高了数据处理效率与准确性。
此外，混元DiT模型的LoRA小规模数据集训练方案也同期开源，为用户在资源有限的情况下进行模型训练提供了新的选择。同时，可控制插件ControlNet的加入，为用户提供了更加精细化的模型控制，满足了多样化的需求。
业界专家表示，腾讯此次开放训练代码及新插件，将极大地推动中文自然语言处理领域的发展，并有望为用户带来更加丰富、智能的应用体验。未来，随着技术的不断进步，中文自然语言处理将在更多领域得到广泛应用。
目前，腾讯混元正积极拥抱开源生态，期待通过技术共享推动整个行业的共同发展。
英语如下：
News Title: “Tencent Opens Source the Mixed Precision Generation Model: Chinese Native Training Code and LoRA, ControlNet Plugins Unveiled”
Keywords: 1. Tencent Open Source
News Content: **Tencent Launches the Mixed Precision Generation Model with Open Source Training Code and New Plugins**
Recently, Tencent announced its latest natural language processing innovation – the open source training code of the Mixed Precision Generation Model. At the same time, the highly anticipated LoRA and ControlNet plugins were also openly sourced. The launch of this technology marks a new milestone in the field of Chinese natural language processing.
It is understood that the open source training code of the Mixed Precision Generation Model (简称“Mixed DiT Model”) provides a more convenient user experience. As a pure Chinese native model, it can directly use Chinese data and labels during training, without the need for complicated data translation processes. This innovation not only simplifies the operation process but also greatly improves data processing efficiency and accuracy.
In addition, the LoRA small-scale dataset training scheme of the Mixed DiT Model was also openly sourced at the same time, providing users with a new choice for model training under limited resources. Meanwhile, the addition of the controllable plugin ControlNet provides users with more refined model control and meets diversified needs.
Industry experts said that Tencent’s open source training code and new plugins will greatly promote the development of the Chinese natural language processing field and is expected to bring richer and more intelligent application experiences to users. In the future, with the continuous progress of technology, Chinese natural language processing will be widely used in more fields.
Currently, Tencent Mixed is actively embracing the open source ecosystem and looking forward to promoting the common development of the entire industry through technology sharing.
【来源】https://ai-bot.cn/go/?url=aHR0cHM6Ly9tcC53ZWl4aW4ucXEuY29tL3MvSkZ4Ukx2blROekszQnlhcUhFUVNPZw%3D%3D
Views: 3