{"id":174772,"date":"2025-10-06T11:48:41","date_gmt":"2025-10-06T11:48:41","guid":{"rendered":"https:\/\/teknomers.com\/en\/we-have-a-new-champion-in-generating-images-with-ai-and-hes-not-american-hes-chinese\/"},"modified":"2025-10-06T11:48:41","modified_gmt":"2025-10-06T11:48:41","slug":"we-have-a-new-champion-in-generating-images-with-ai-and-hes-not-american-hes-chinese","status":"publish","type":"post","link":"https:\/\/teknomers.com\/en\/we-have-a-new-champion-in-generating-images-with-ai-and-hes-not-american-hes-chinese\/","title":{"rendered":"We have a new champion in generating images with AI, and he’s not American; he’s Chinese."},"content":{"rendered":"\n

Tencent Launches Hunyuan Image 3.0: A Game-Changer in AI-Generated Imagery<\/h2>\n

\u00a0Tencent\u00a0 has just unveiled its latest model, \u00a0Hunyuan Image 3.0\u00a0, capable of generating \u00a0images from text prompts\u00a0. Traditionally, proprietary models have been at the forefront of this creative technology, but recent benchmarks indicate that \u00a0Tencent’s model\u00a0 has outperformed its competitors, including tech giants like Google and OpenAI.<\/p>\n

According to the LMARENA classification<\/a>, \u00a0Hunyuan Image 3.0\u00a0 has surpassed models such as \u00a0Gemini 2.5 Flash\u00a0\u2014popularly known as \u00a0Nano Banana\u00a0\u2014alongside other proprietary systems like \u00a0GPT-IMAGE-1\u00a0, \u00a0FLUX-1-KONTEXT-MAX\u00a0, and \u00a0QWEN-IMAGE\u00a0. This remarkable leap forward establishes Tencent as a formidable player in the \u00a0AI-generated imagery\u00a0 sector.<\/p>\n

<\/p>\n

\n
\n
\n \"Screen
\n The Tencent image model has already managed to overcome its competitors according to LM Arena experts.<\/span>\n <\/div>\n<\/p><\/div>\n<\/div>\n

\u00a0Blind Vote System\u00a0: The *LMA Arena classification* operates on a \u00a0blind voting system\u00a0, allowing users to select their favorite images without knowing which model generated them. Preliminary results show that this Tencent model outperformed all its rivals, including Google’s Nano Banana. However, it is crucial to note that the results are still classified as \u00a0preliminary\u00a0 due to the model\u2019s short availability period.<\/p>\n

\u00a0How Hunyuan Image 3.0 Works\u00a0: Tencent has revealed that they have employed a new \u00a0diffusion architecture\u00a0, utilizing dual encoders: a \u00a0multimodal LLM\u00a0 and another that enhances language character recognition. Additionally, the \u00a0Reinforcement Learning from Human Feedback (RLHF)\u00a0 technique refines the images produced. This dual programming allows the production of high-quality images while minimizing resource consumption.<\/p>\n

<\/p>\n

\n
\n
\n
\n \"Nano\n <\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/div>\n

\u00a0Open Weights and Commercial License\u00a0: Hunyuan Image 3.0 stands out by sharing its code on GitHub<\/a>. \u00a0Tencent\u00a0 offers a surprisingly permissive license that allows the model to be utilized for \u00a0commercial and professional purposes\u00a0. This open-source approach marks a significant shift in how advanced technologies are distributed and utilized.<\/p>\n

<\/p>\n

\u00a0Cost Structure\u00a0: While the model can be tested for free on its official project website<\/a>, the credits system reveals a less economical pricing structure: for \u00a0$8\u00a0 per month, users acquire \u00a0500 credits\u00a0, allowing approximately \u00a050 images\u00a0 to be generated at \u00a0$0.16 each\u00a0. Comparatively, Nano Banana’s images cost only \u00a0$0.039\u00a0, making it a more budget-friendly option. Nonetheless, alternatives for testing, such as Hugging Face “Spaces”<\/a>, and the option for an API key through Tencent Cloud<\/a>, provide variety for potential users.<\/p>\n

<\/p>\n

\n
\n
\n \"Screen
\n \"Hunyuan
\n <\/a>\n <\/div>\n<\/div>\n

\u00a0Navigating the AI Landscape\u00a0: Although the Tencent model is a notable contender, Nano Banana continues to hold its ground as a top choice for users seeking to edit images. It has evolved into an alternative to traditional software like \u00a0Photoshop\u00a0. Users can input their desired edits directly to the AI rather than working through conventional editing tools.<\/p>\n

<\/p>\n

\u00a0Conversational Models\u00a0: While Hunyuan Image 3.0 offers impressive capabilities, it appears more focused on single prompts compared to the conversational interface of Gemini. Competitors like \u00a0Alibaba\u00a0 with \u00a0Qwen-Image-Editor\u00a0 emphasize user interaction, allowing for dialogue with images for real-time modifications. This approach contrasts with Tencent’s more straightforward prompt-based system.<\/p>\n

<\/p>\n

\u00a0The Industry Implication\u00a0: The emergence of Hunyuan Image 3.0 suggests that performance in generating images, previously thought to be dominated by proprietary models, can now rival or even exceed \u00a0open-source alternatives\u00a0. This move reflects a broader trend in the industry, emphasizing how Chinese companies are challenging the traditional, closed-off philosophy much admired in the \u00a0American tech sector\u00a0. <\/p>\n

Image insights can be found in their latest developments and interactive demonstrations via Hunyuan<\/strong>\u2019s Twitter. With Hunyuan Image 3.0<\/strong>, we are reminded that innovation is multidimensional, and the future of AI-generated imagery might be exponentially more inclusive and accessible than ever before.<\/p>\n


\n
General News – 2<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"

Tencent Launches Hunyuan Image 3.0: A Game-Changer in AI-Generated Imagery \u00a0Tencent\u00a0 has just unveiled its latest model, \u00a0Hunyuan Image 3.0\u00a0, capable of generating \u00a0images from text prompts\u00a0. Traditionally, proprietary models have been at the forefront of this creative technology, but recent benchmarks indicate that \u00a0Tencent’s model\u00a0 has outperformed its competitors, including tech giants like Google […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[36399],"tags":[2381,1828,2394,39020,10811,5250],"class_list":["post-174772","post","type-post","status-publish","format-standard","hentry","category-technology","tag-american","tag-champion","tag-chinese","tag-generating","tag-hes","tag-images"],"_links":{"self":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts\/174772","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/comments?post=174772"}],"version-history":[{"count":0,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts\/174772\/revisions"}],"wp:attachment":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/media?parent=174772"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/categories?post=174772"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/tags?post=174772"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}