Lovart
<h1>Lovart 全球首个设计 Agent 体验</h1>
<section><span data-pm-slice="1 1 [&quot;para&quot;,{&quot;tagName&quot;:&quot;p&quot;,&quot;attributes&quot;:{&quot;style&quot;:&quot;font-style: normal;font-variant-ligatures: normal;font-variant-caps: normal;font-weight: 400;letter-spacing: normal;orphans: 2;text-align: start;text-indent: 0px;text-transform: none;widows: 2;word-spacing: 0px;-webkit-text-stroke-width: 0px;white-space: normal;text-decoration-thickness: initial;text-decoration-style: initial;text-decoration-color: initial;font-size: 15px;line-height: 2em;font-family: \&quot;PingFang SC\&quot;, system-ui, -apple-system, BlinkMacSystemFont, \&quot;Helvetica Neue\&quot;, \&quot;Hiragino Sans GB\&quot;, \&quot;Microsoft YaHei UI\&quot;, \&quot;Microsoft YaHei\&quot;, Arial, sans-serif;color: rgb(31, 35, 41);margin: 0px 0px 4px;word-break: break-all;min-height: 20px;&quot;},&quot;namespaceURI&quot;:&quot;http://www.w3.org/1999/xhtml&quot;},&quot;node&quot;,{&quot;tagName&quot;:&quot;span&quot;,&quot;attributes&quot;:{&quot;style&quot;:&quot;color: rgb(31, 35, 41);&quot;},&quot;namespaceURI&quot;:&quot;http://www.w3.org/1999/xhtml&quot;}]">Lovart 的三个特点:</span></section>
<h3>一、全链路设计和执行,一句话搞定</h3>
<p>以前的文生图工具,它们所提供的任务是&ldquo;生成图片&rdquo;这一环。</p>
<p>而设计 Agent,则像一位&ldquo;设计执行官&rdquo;,覆盖从创意拆解到专业交付的整个视觉流程。</p>
<p>从意图拆解 &rarr; 任务链 &rarr; 最后成品,一句话全搞定。</p>
<p>单次可以执行上百步,最多可以一次产出40张设计成品图。</p>
<p>比如,我可以让它根据我家猫的照片做一套银渐层表情包,它会自己设计表情,挨个生图,还可以全部编辑,一键做出所有表情包的黑底渐变色版本。</p>
<p>&nbsp;</p>
<section><img class="qnt-img-img qnr-img-lazy-load-img" src="https://inews.gtimg.com/om_bt/ORPkpWhOF_KMtPcoydMfk6zoBDmmg3KCqFlEd7j2vFDVIAA/1000" alt="图片" data-src="https://inews.gtimg.com/om_bt/ORPkpWhOF_KMtPcoydMfk6zoBDmmg3KCqFlEd7j2vFDVIAA/1000"></section>
<p>Prompt:</p>
<p>Use the cat in this picture as a template to design a set of 12 silver gradient emoji packs. Style 3D cartoon, to express the most popular and trendy emoji this year.</p>
<p>&nbsp;</p>
<p>再比如,为宣传智能家居扬声器的 30 秒视频广告创建一个 6 帧故事板。</p>
<p>它会先根据需求,写出故事,然后开始进行角色设计。</p>
<p>完成之后,会进行关键场景的设计。</p>
<p>在最后把场景填充到故事板里,为每个画面配上光线、摄影机移动等信息。</p>
<p><img class="qnt-img-img qnr-img-lazy-load-img" src="https://inews.gtimg.com/om_bt/G2tapNA9henwcYDaY6W7SIIYNY93t5vys-950CiJWwTkcAA/0" alt="图片" data-src="https://inews.gtimg.com/om_bt/G2tapNA9henwcYDaY6W7SIIYNY93t5vys-950CiJWwTkcAA/0"></p>
<p data-pm-slice="0 0 []">Prompt:</p>
<p>Create a 6-frame storyboard for a 30-second video ad promoting a smart home speaker.</p>
<p>Scene progression:</p>
<p>Modern living room, family interacting</p>
<p>Voice command to the speaker</p>
<p>Music starts playing</p>
<p>Speaker answers a question</p>
<p>Lighting adjusts automatically</p>
<p>Close-up of product and brand tagline</p>
<p>Style: clean, semi-realistic illustration, soft light.</p>
<p>&nbsp;</p>
<h3>二、图像视频音乐,自由调度</h3>
<p>只有具备了充足的工具箱,<span data-pm-slice="1 1 [&quot;para&quot;,{&quot;tagName&quot;:&quot;p&quot;,&quot;attributes&quot;:{&quot;style&quot;:&quot;font-style: normal;font-variant-ligatures: normal;font-variant-caps: normal;font-weight: 400;letter-spacing: normal;orphans: 2;text-align: start;text-indent: 0px;text-transform: none;widows: 2;word-spacing: 0px;-webkit-text-stroke-width: 0px;white-space: normal;text-decoration-thickness: initial;text-decoration-style: initial;text-decoration-color: initial;font-size: 15px;line-height: 2em;font-family: \&quot;PingFang SC\&quot;, system-ui, -apple-system, BlinkMacSystemFont, \&quot;Helvetica Neue\&quot;, \&quot;Hiragino Sans GB\&quot;, \&quot;Microsoft YaHei UI\&quot;, \&quot;Microsoft YaHei\&quot;, Arial, sans-serif;color: rgb(31, 35, 41);margin: 0px 0px 4px;word-break: break-all;min-height: 20px;&quot;},&quot;namespaceURI&quot;:&quot;http://www.w3.org/1999/xhtml&quot;},&quot;node&quot;,{&quot;tagName&quot;:&quot;span&quot;,&quot;attributes&quot;:{&quot;style&quot;:&quot;color: rgb(31, 35, 41);&quot;},&quot;namespaceURI&quot;:&quot;http://www.w3.org/1999/xhtml&quot;}]">Agent 才能更全面地完成任务。</span></p>
<p>Lovart 集成了一系列前沿的 AI 模型,比如 GPT image-1, Flux pro, OpenAI-o3, Gemini Imagen 3, Kling AI, Tripo AI, Suno AI等。</p>
<p>设计师可以在一个统一的界面内完成设计、编辑与生成,再也不必在多个软件和模型之间来回切换。</p>
<p>我让它设计一个 "VIVIDPIX "的新生活科技品牌,并创建跨媒体发布活动。</p>
<p>它不仅设计了logo、海报、周边,还用可灵制作了3D版本的logo动画,用suno制作了bgm。</p>
<p>由于所有的模型都拥有相同的 Context 上下文,配合起来丝滑无缝。</p>
<section><img class="qnt-img-img qnr-img-lazy-load-img" src="https://inews.gtimg.com/om_bt/GCAo0LAx9FNM5OxMeFJpzbuSnyZ4legCKsC_oEMyW6mHcAA/0" alt="图片" data-src="https://inews.gtimg.com/om_bt/GCAo0LAx9FNM5OxMeFJpzbuSnyZ4legCKsC_oEMyW6mHcAA/0"></section>
<p>Prompt:</p>
<p data-pm-slice="0 0 []">Prompt:</p>
<p>Create a cross-media launch campaign for a new lifestyle tech brand called &ldquo;VIVIDPIX&rdquo;.</p>
<p>Visuals: Design two main brand posters in different styles (futuristic minimalism and playful retro), including editable text layers and logo integration.</p>
<p>Video/3D: Generate a dynamic 10-second video loop of the brand logo morphing in 3D space, with smooth, modern animation, suitable for website and social sharing. Leverage Kling or equivalent video/animation AI tools.</p>
<p>Music: Compose an original 15-second catchy theme music that matches the brand&rsquo;s energetic and optimistic tone. Use Suno or other advanced AI music models.</p>
<p>Packaging: Output a box and product mockup, harmonizing visuals across all touchpoints.</p>
<p>Canvas: Bring all assets together onto a single editable Lovart canvas, allowing for integrated fine-tuning and end-to-end brand asset delivery.</p>
<p>All elements&mdash;visuals, video, music, packaging&mdash;should have a cohesive color palette and brand personality, and be ready to use for a multi-platform product launch.</p>
<p>&nbsp;</p>
<h3>三、生成结果自动分图层,随心修改</h3>
<p>在海报的生成中,最大的痛点是文字后期的编辑,Lovart 很好地解决了这一痛点。</p>
<p>举例来说,我这里要为一种新的植物蛋白饮料设计一张宣传海报。</p>
<p>一分钟后,它就生成了左侧的标准海报,同时在右侧附上了图文分离的版本。</p>
<p>现在我可以自由修改文字的内容、大小、颜色,真的太方便了。</p>
<section><img class="qnt-img-img qnr-img-lazy-load-img" src="https://inews.gtimg.com/om_bt/GzxWhUIVamAgAR50Ayk23PIkiTk6CwOeHZ3-GL7UhzA9EAA/0" alt="图片" data-src="https://inews.gtimg.com/om_bt/GzxWhUIVamAgAR50Ayk23PIkiTk6CwOeHZ3-GL7UhzA9EAA/0"></section>
<p data-pm-slice="0 0 []">Prompt:</p>
<p>Design a promotional poster for a new plant-based protein drink.</p>
<p>Style: vibrant and energetic, modern editorial layout.</p>
<p>Color palette: green, cream, and orange.</p>
<p>Text: &ldquo;Fuel Your Day, Naturally&rdquo;.</p>
<p>Aspect ratio: 3:4, for Instagram feed.</p>
<p>Include editable text layers.</p>
<p>&nbsp;</p>
<h2>使用小技巧:</h2>
<p>在使用 Lovart 的过程中也发现了一些小心得,在这里分享给大家:</p>
<ol class="list-paddingleft-1">
<li>
<section>1.&nbsp;Lovart 内置了很多图像、视频、音乐模型,但并不是每个模型都对中文支持够好,所以使用英文进行提示词会获得最佳体验。</section>
</li>
<li>
<section>2.&nbsp;设计的起点是需求,只有把需求描述好,Agent 才能设计好,所以建议在发布任务之前,先和 AI 聊一下需求,让 AI 输出最后的提示词。提示词里可以包含风格、颜色、文字、比例等结构化信息,结构化信息本身是对复杂需求的一种拆解,这样写出来的提示词,Lovart 在执行的时候效果也最好。</section>
</li>
<li>
<section>3.&nbsp;如果需要独立的文字图层,只需要在指令中加上这句话即可: Include editable text layers.</section>
</li>
</ol>
<p>&nbsp;</p>
<h2>结语</h2>
<p>Lovart 的出现,代表 AI 在设计领域的角色,正在从单纯的&ldquo;图像生成器&rdquo;向更深层次的&ldquo;设计执行官&rdquo;转变。</p>
<p>它试图解决的,不仅仅是&ldquo;画出一张图&rdquo;,更是&ldquo;完成一个设计任务&rdquo;的问题。</p>
<p>&nbsp;</p>
<p>有人曾问我,4o 生图的出现是否会干掉图像的工作流。</p>
<section>我所看到的情况,不是干掉,而是解放。</section>
<section>这种解放,将图像工作流推向设计 Agent 的新范式。</section>
<section></section>
<section>技术的发展总是在不断挑战我们的想象边界。</section>
<p>当 AI Agent 开始尝试理解并执行更复杂的设计流程时,</p>
<p>我们与 AI ,<span data-pm-slice="0 0 []">也将迎来全新的</span>协作方式。</p>