{"id":10363,"date":"2024-05-15T09:29:53","date_gmt":"2024-05-15T01:29:53","guid":{"rendered":"https:\/\/www.1ai.net\/?p=10363"},"modified":"2024-05-15T09:29:53","modified_gmt":"2024-05-15T01:29:53","slug":"%e8%85%be%e8%ae%af%e6%b7%b7%e5%85%83%e6%96%87%e7%94%9f%e5%9b%be%e5%a4%a7%e6%a8%a1%e5%9e%8b%e5%af%b9%e5%a4%96%e5%bc%80%e6%ba%90%ef%bc%9a%e6%90%ad%e8%bd%bd%e9%a6%96%e4%b8%aa%e4%b8%ad%e8%8b%b1%e5%8f%8c","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/10363.html","title":{"rendered":"Tencent's Hunyuan Wenshengtu model is open source: equipped with the first Chinese-English bilingual DiT architecture, free for commercial use"},"content":{"rendered":"

Tencent<\/a>Announced its Hunyuan WenshengtuLarge Model<\/a>Upgrade and openOpen Source<\/a><\/strong>, which has been released on Hugging Face and Github, contains a complete model including model weights, inference code, model algorithms, etc., which can be used by enterprises and individual developersFree for commercial use<\/strong>.<\/p>\n

\"a1cf56fd-067d-4e75-96ec-581c81c23d8f\"<\/p>\n

\u25b2 Hunyuan Wenshengtu effect<\/p>\n

The upgraded Hunyuan Wenshengtu model uses the same DiT architecture as Sora. Tencent said that Hunyuan DiT isThe first bilingual DiT framework in Chinese and English<\/strong>Hunyuan DiT is a text-to-image generation model based on the Diffusion transformer. This model has the ability to understand Chinese and English in fine-grained terms. Hunyuan DiT can have multiple rounds of dialogue with users and generate and improve images based on the context. This is also the first in the industry.The first native Chinese<\/strong>The DiT architecture Wenshengtu open source model supports bilingual input and understanding in Chinese and English, with 1.5 billion parameters.<\/p>\n

\"2c14d3ed-7e95-416b-a179-57220d333adb\"<\/p>\n

Running this model requiresCUDA-enabled NVIDIA GPU<\/strong>, required to run Hunyuan DiT aloneMinimum video memory is 11GB<\/strong>, running DialogGen (a text-to-image multimodal interactive dialogue system launched by Tencent) and Hunyuan DiT at the same timeAt least 32GB of video memory is required<\/strong>, Tencent said they have tested Nvidia's V100 and A100 GPUs on Linux.<\/p>\n

\"af27aed4-0bf1-435e-aa9d-cffac6dc4f4b\"<\/p>\n

\u25b2 Configuration requirements<\/p>\n

According to previous reports, the results of the first official "large model standard conformity assessment" in China were announced. Tencent's Hunyuan large model became the first batch of domestic large models to pass the assessment. Other large models that passed the assessment included Alibaba Tongyi Qianwen, 360 Zhinao and Baidu Wenxin Yiyan.<\/p>","protected":false},"excerpt":{"rendered":"

Tencent has announced that it has upgraded and open-sourced its hybrid graph model, which is now available on Hugging Face and Github, and includes a complete model with model weights, inference code, and model algorithms that can be commercialized for free by both enterprise and individual developers. The upgraded hybrid Wenshengtu model adopts the same DiT architecture as Sora, and Tencent said that hybrid DiT is the first bilingual DiT architecture. The DiT is a text-to-image generation model based on Diffusion transformer, which has the ability to understand Chinese and English at a fine-grained level. The DiT is able to conduct multiple rounds of dialog with the user to generate and refine images based on the context. It is also<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[216,219,323],"collection":[],"class_list":["post-10363","post","type-post","status-publish","format-standard","hentry","category-news","tag-216","tag-219","tag-323"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/10363","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=10363"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/10363\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=10363"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=10363"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=10363"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=10363"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}