{"id":18890,"date":"2026-09-06T01:24:04","date_gmt":"2026-09-06T05:24:04","guid":{"rendered":"https:\/\/wp.glbgpt.com\/?p=18890"},"modified":"2026-09-06T01:28:39","modified_gmt":"2026-09-06T05:28:39","slug":"wan-2-5","status":"publish","type":"post","link":"https:\/\/wp.glbgpt.com\/jp\/hub\/wan-2-5","title":{"rendered":"Wan 2.5: Alibaba&#8217;s AI Video Generator Explained"},"content":{"rendered":"<style>\n.wan25-article{--ink:#293238;--muted:#687178;--sage:#5f7567;--sage-dark:#496056;--sage-soft:#edf2ee;--gold:#ad7c28;--gold-soft:#fff7e8;--rose:#985f58;--rose-soft:#fbf3f1;--line:#d8dedb;--paper:#fff;--stone:#f5f6f4;color:var(--ink);font-size:16px;line-height:1.7}\n.wan25-article *{box-sizing:border-box}.wan25-article h2,.wan25-article h3{letter-spacing:0}.wan25-article h2{margin-top:42px;padding-bottom:9px;border-bottom:1px solid var(--line)}.wan25-article h3{margin-top:28px;color:#384940}.wan25-article p{margin-top:0;margin-bottom:18px}.wan25-article a{color:var(--sage-dark);text-underline-offset:2px}.wan25-article li{margin-bottom:7px}.wan25-article li::marker{color:var(--gold)}\n.wan25.quick-answer{border-left:4px solid var(--gold);background:var(--gold-soft);padding:18px 20px;margin:22px 0}.wan25.quick-answer p{margin:0}\n.wan25.toc{border-left:3px solid var(--sage);padding:4px 0 4px 18px;margin:28px 0}.wan25.toc strong{color:var(--sage-dark)}.wan25.toc ul{columns:2;column-gap:34px;margin:10px 0 0;padding-left:20px}.wan25.toc li{break-inside:avoid}\n.wan25.table-wrap{overflow-x:auto;margin:24px 0;border:1px solid var(--line)}.wan25.table-wrap table{width:100%;min-width:760px;border-collapse:separate;border-spacing:0}.wan25.table-wrap th{background:var(--sage);color:#fff}.wan25.table-wrap th,.wan25.table-wrap td{padding:12px 14px;border-bottom:1px solid var(--line);text-align:left;vertical-align:top}.wan25.table-wrap tr:last-child td{border-bottom:0}.wan25.table-wrap tbody tr:nth-child(even){background:var(--stone)}\n.wan25.fact-check{border-left:4px solid var(--rose);background:var(--rose-soft);padding:16px 18px;margin:22px 0}.wan25.prompt-formula{border-top:1px solid var(--line);border-bottom:1px solid var(--line);padding:14px 0;margin:18px 0}.wan25.prompt-formula code{display:block;padding:12px 14px;background:var(--stone);white-space:normal}\n.wan25.cta{border:1px solid var(--line);border-left:4px solid var(--sage);background:var(--sage-soft);padding:20px;margin:30px 0}.wan25.cta p:first-child{margin-top:0}.wan25.cta .wp-block-buttons{margin:4px 0 0}.wan25.cta .wp-block-button__link{border-radius:6px;background:var(--sage);color:#fff;text-decoration:none;font-weight:700;padding:11px 18px}.wan25.cta .wp-block-button__link:hover{background:var(--sage-dark)}\n.wan25-article .wp-block-image{margin:28px 0}.wan25-article .wp-block-image img{border:1px solid var(--line)}.wan25-article .wp-block-image figcaption{color:var(--muted);font-size:13px}\n@media(max-width:760px){.wan25.toc ul{columns:1}.wan25.quick-answer,.wan25.fact-check,.wan25.cta{padding:16px}.wan25.table-wrap table{min-width:680px}}\n<\/style>\n\n\n\n<div class=\"wp-block-group wan25-article is-layout-constrained wp-block-group-is-layout-constrained\">\n<div class=\"wp-block-group wan25 quick-answer is-layout-constrained wp-block-group-is-layout-constrained\">\n<p class=\"wp-block-paragraph\"><strong>\u7c21\u5358\u306a\u7b54\u3048\uff1a<\/strong> Wan 2.5 is Alibaba&#8217;s hosted AI video model for text-to-video and image-to-video generation. Its official Model Studio endpoints can produce 5- or 10-second MP4 clips at 480P, 720P, or 1080P, with synchronized audio generated alongside the visuals. The model remains in Alibaba Cloud&#8217;s documentation, although it is no longer the newest release in the Wan family.<\/p>\n<\/div>\n\n\n\n<p class=\"wp-block-paragraph\">Much of the confusion around Wan 2.5 comes from its relationship to Alibaba&#8217;s open-source Wan models. They belong to the same family, but the official Wan 2.5 video endpoints were released as hosted Preview services. Alibaba has not published verified Wan 2.5 weights for local ComfyUI use, so access, licensing, and setup differ from open-weight releases such as Wan 2.2.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">For readers who prefer keeping chat, research, image, video, and agent tools in one place, <a href=\"https:\/\/www.glbgpt.com\/home?inviter=hub_popup&amp;login=1\">\u30b0\u30ed\u30fc\u30d0\u30ebGPT<\/a> provides an all-in-one workspace under one account. It can be a practical way to check whether Wan 2.5 or a newer Wan model is available without moving between several provider sites. Because availability changes, check the live Video picker before starting a project.<\/p>\n\n\n\n<div class=\"wp-block-group is-layout-constrained wp-block-group-is-layout-constrained\">\n<figure class=\"wp-block-image size-large\"><a href=\"https:\/\/www.glbgpt.com\/video-generator\/wan-3-0?inviter=hub_content_wan27&amp;login=1\"><img alt=\"\" fetchpriority=\"high\" decoding=\"async\" width=\"1024\" height=\"633\" src=\"https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/08\/wan-3-0-1024x633.png\" class=\"wp-image-18449\" srcset=\"https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/08\/wan-3-0-1024x633.png 1024w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/08\/wan-3-0-300x185.png 300w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/08\/wan-3-0-767x474.png 767w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/08\/wan-3-0-18x11.png 18w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/08\/wan-3-0-1536x949.png 1536w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/08\/wan-3-0-2048x1265.png 2048w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/a><\/figure>\n\n\n\n<div class=\"wp-block-buttons is-content-justification-center is-layout-flex wp-container-core-buttons-is-layout-3e41869c wp-block-buttons-is-layout-flex\">\n<div class=\"wp-block-button is-style-outline is-style-outline--1\"><a class=\"wp-block-button__link has-white-color has-vivid-purple-background-color has-text-color has-background has-link-color wp-element-button\" href=\"https:\/\/www.glbgpt.com\/video-generator\/wan-3-0?inviter=hub_content_wan27&amp;login=1\">\u4eca\u3059\u3050 GlobalGPT \u3067 Wan 3.0 \u3092\u8a66\u3057\u3066\u307f\u307e\u3057\u3087\u3046<\/a><\/div>\n<\/div>\n<\/div>\n\n\n\n<div class=\"wp-block-group wan25 toc is-layout-constrained wp-block-group-is-layout-constrained\">\n<p class=\"wp-block-paragraph\"><strong>\u76ee\u6b21<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a href=\"#what-is-wan-2-5\">What is Wan 2.5 AI?<\/a><\/li>\n\n\n\n<li><a href=\"#wan-2-5-specs\">Wan 2.5 specifications<\/a><\/li>\n\n\n\n<li><a href=\"#wan-2-5-features\">What can Wan 2.5 do?<\/a><\/li>\n\n\n\n<li><a href=\"#wan-2-5-image-to-video\">Wan 2.5 image to video<\/a><\/li>\n\n\n\n<li><a href=\"#how-to-use-wan-2-5\">How to use Wan 2.5<\/a><\/li>\n\n\n\n<li><a href=\"#is-wan-2-5-free\">Is Wan 2.5 free?<\/a><\/li>\n\n\n\n<li><a href=\"#wan-2-5-open-source\">Is Wan 2.5 open source?<\/a><\/li>\n\n\n\n<li><a href=\"#wan-2-5-comfyui\">Can Wan 2.5 run in ComfyUI?<\/a><\/li>\n\n\n\n<li><a href=\"#wan-model-comparison\">Wan 2.5 vs newer Wan models<\/a><\/li>\n\n\n\n<li><a href=\"#wan-2-5-prompts\">Prompt checklist<\/a><\/li>\n\n\n\n<li><a href=\"#wan-2-5-faq\">\u3088\u304f\u3042\u308b\u3054\u8cea\u554f<\/a><\/li>\n<\/ul>\n<\/div>\n\n\n\n<h2 id=\"what-is-wan-2-5\" class=\"wp-block-heading\">What Is Wan 2.5 AI?<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Wan 2.5 is part of Alibaba&#8217;s Wan visual-generation family. Alibaba previewed four Wan 2.5 models at Apsara Conference 2025, including video generation, image generation, and image editing models. The company described the video models as using a jointly trained text, audio, and visual architecture designed to improve prompt following and synchronize sound with the generated picture.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The timing explains the original headline. In its <a href=\"https:\/\/www.alibabacloud.com\/blog\/alibaba-cloud-unveils-strategic-roadmaps-for-the-next-generation-ai-innovations_602560\">September 2025 Wan 2.5 announcement<\/a>, Alibaba positioned the model alongside its larger full-stack AI push. Wan 2.5 was not simply another silent video generator. Native audio was the feature that moved it into the same conversation as premium hosted video systems.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full is-resized\"><img decoding=\"async\" width=\"692\" height=\"285\" src=\"https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/01-wan25-official-launch.webp\" alt=\" Wan 2.5 AI\u52d5\u753b\u751f\u6210\u30c4\u30fc\u30eb\uff1a\u753b\u50cf\u304b\u3089\u52d5\u753b\u3078\u306e\u30e2\u30fc\u30b7\u30e7\u30f3\u5909\u63db\u6a5f\u80fd\u3068\u97f3\u58f0\u306e\u540c\u671f\u6a5f\u80fd\u3092\u642d\u8f09\" class=\"wp-image-18891\" style=\"aspect-ratio:2.418127218127218;width:800px;height:auto\" srcset=\"https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/01-wan25-official-launch.webp 692w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/01-wan25-official-launch-300x124.webp 300w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/01-wan25-official-launch-18x7.webp 18w\" sizes=\"(max-width: 692px) 100vw, 692px\" \/><figcaption class=\"wp-element-caption\">Alibaba introduced four Wan2.5 Preview models at Apsara Conference 2025. Source:\u00a0<a href=\"https:\/\/www.alibabacloud.com\/blog\/alibaba-cloud-unveils-strategic-roadmaps-for-the-next-generation-ai-innovations_602560\">Alibaba Cloud<\/a>.<\/figcaption><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">By September 2026, however, Alibaba&#8217;s product line had moved forward. The official Model Studio catalog still lists Wan 2.5 Preview, while newer Wan 2.7 and Wan 3.0 endpoints offer longer clips or broader reference controls. If you are researching the current family rather than this specific version, start with the <a href=\"https:\/\/www.glbgpt.com\/hub\/wan-3-0\/\">Wan 3.0 release and feature guide<\/a>.<\/p>\n\n\n\n<h2 id=\"wan-2-5-specs\" class=\"wp-block-heading\">Wan 2.5 Specifications<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">\u306b\u3064\u3044\u3066 <a href=\"https:\/\/www.alibabacloud.com\/help\/en\/model-studio\/video-generate-edit-model\/\">Alibaba Cloud Model Studio catalog<\/a>, updated September 2, 2026, lists two Wan 2.5 Preview endpoints:<\/p>\n\n\n\n<figure class=\"wp-block-table wan25 table-wrap\"><table><thead><tr><th>\u30e2\u30c7\u30ebID<\/th><th>Input mode<\/th><th>\u30aa\u30fc\u30c7\u30a3\u30aa<\/th><th>\u6c7a\u8b70<\/th><th>\u671f\u9593<\/th><th>\u51fa\u529b<\/th><\/tr><\/thead><tbody><tr><td><code>wan2.5-t2v-preview<\/code><\/td><td>\u30c6\u30ad\u30b9\u30c8\u304b\u3089\u30d3\u30c7\u30aa\u3078<\/td><td>Audio sync<\/td><td>480P, 720P, 1080P<\/td><td>5 or 10 seconds<\/td><td>30 fps, MP4<\/td><\/tr><tr><td><code>wan2.5-i2v-preview<\/code><\/td><td>\u753b\u50cf\u304b\u3089\u52d5\u753b\u3078<\/td><td>Audio sync<\/td><td>480P, 720P, 1080P<\/td><td>5 or 10 seconds<\/td><td>30 fps, MP4<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<figure class=\"wp-block-image size-large\"><img decoding=\"async\" width=\"1024\" height=\"228\" src=\"https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/02-wan25-model-specs-1-1024x228.webp\" alt=\"Alibaba Cloud announcement describing Wan 2.5 Preview\" class=\"wp-image-18893\" srcset=\"https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/02-wan25-model-specs-1-1024x228.webp 1024w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/02-wan25-model-specs-1-300x67.webp 300w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/02-wan25-model-specs-1-765x170.webp 765w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/02-wan25-model-specs-1-18x4.webp 18w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/02-wan25-model-specs-1.webp 1260w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><figcaption class=\"wp-element-caption\">Alibaba Cloud lists two Wan 2.5 Preview endpoints with synchronized audio. Source:\u00a0<a href=\"https:\/\/www.alibabacloud.com\/help\/en\/model-studio\/video-generate-edit-model\/\">Alibaba Cloud Model Studio documentation<\/a>.<\/figcaption><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">\u305d\u306e\u8a00\u8449 <em>\u30d7\u30ec\u30d3\u30e5\u30fc<\/em> is not decorative. It is part of both model IDs and signals a hosted product endpoint, not a downloadable checkpoint. It also means you should verify access and billing in the platform you plan to use instead of trusting an old bookmarked route.<\/p>\n\n\n\n<h2 id=\"wan-2-5-features\" class=\"wp-block-heading\">What Can the Wan 2.5 AI Video Generator Do?<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\">Generate video and sound together<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Wan 2.5&#8217;s defining upgrade was synchronized audio. Alibaba said the model could create high-fidelity audio with the video rather than requiring creators to build the sound layer in a separate tool. For a short dialogue, product shot, or atmospheric scene, that can remove a full editing step.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Turn prompts into short 1080P clips<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">The text-to-video endpoint accepts a written scene description and returns a 5- or 10-second clip. A useful prompt should describe the subject, action, camera, setting, visual treatment, and sound. That structure is more reliable than stacking vague quality words such as &#8220;cinematic&#8221; and &#8220;stunning.&#8221;<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Animate a still image<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">The image-to-video endpoint gives the model a visual starting point. This is useful when the exact product, character, composition, or color palette matters more than open-ended invention. For a broader view of the market, the <a href=\"https:\/\/www.glbgpt.com\/hub\/best-ai-video-generator\/\">AI\u52d5\u753b\u751f\u6210\u30c4\u30fc\u30eb\u306e\u6bd4\u8f03\u30e9\u30f3\u30ad\u30f3\u30b0<\/a> shows how different tools approach quality, control, and workflow.<\/p>\n\n\n\n<h2 id=\"wan-2-5-image-to-video\" class=\"wp-block-heading\">Wan 2.5 Image to Video: A Practical Workflow<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Start with a clean source image. A frame with one obvious subject, readable separation from the background, and enough room for movement gives the model fewer conflicts to solve. If the source already contains distorted hands, unreadable text, or impossible geometry, motion usually makes those flaws more visible.<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Choose the first frame.<\/strong> Use the image that best represents the subject and final art direction.<\/li>\n\n\n\n<li><strong>Describe motion, not appearance alone.<\/strong> Tell the model what the subject does and how fast it happens.<\/li>\n\n\n\n<li><strong>Direct the camera.<\/strong> Use one clear move such as a slow push-in, lateral tracking shot, or locked camera.<\/li>\n\n\n\n<li><strong>Write the sound layer.<\/strong> Describe dialogue, ambience, sound effects, or music only when it helps the scene.<\/li>\n\n\n\n<li><strong>Generate a short draft first.<\/strong> Fix motion and composition before spending more on higher resolution.<\/li>\n<\/ol>\n\n\n\n<p class=\"wp-block-paragraph\">Image-to-video models do not all behave the same way. The <a href=\"https:\/\/www.glbgpt.com\/hub\/kling-ai-image-to-video-guide-for-creators\/\">Kling AI image-to-video workflow<\/a> is a useful comparison when you need a second approach to source-image motion and social video production.<\/p>\n\n\n\n<h2 id=\"how-to-use-wan-2-5\" class=\"wp-block-heading\">How to Use Wan 2.5 in 2026<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\">Option 1: Alibaba Cloud Model Studio<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Model Studio is the official hosted route for the two Wan 2.5 Preview endpoints. It is the clearest choice for developers who need named model IDs, API-controlled jobs, and provider documentation. You will need an Alibaba Cloud account, the correct deployment scope, and enough quota or billing balance for the job.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Option 2: GlobalGPT&#8217;s video workspace<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">GlobalGPT is designed for users who would rather keep several AI workflows under one account and one subscription. Its <a href=\"https:\/\/www.glbgpt.com\/order\">\u516c\u7684\u30d7\u30e9\u30f3\u306e\u30da\u30fc\u30b8<\/a> listed Wan 2.5 among its included video models when checked on September 3, 2026, alongside newer Wan options. The practical benefit is broader capability coverage without maintaining separate subscriptions for every provider.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Open the Video workspace and inspect the live model picker before starting. If Wan 2.5 is missing, do not force an old dedicated URL. Choose Wan 2.7 or Wan 3.0 based on the controls and duration you need. Users comparing the newest family can also review <a href=\"https:\/\/www.glbgpt.com\/hub\/wan-3-0-prime\/\">Wan 3.0 Prime&#8217;s speed and pricing differences<\/a>.<\/p>\n\n\n\n<h2 id=\"is-wan-2-5-free\" class=\"wp-block-heading\">Is Wan 2.5 Free?<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Wan 2.5 is not a freely downloadable open model.<\/strong> It is a hosted Preview service, so usage depends on the provider&#8217;s current quota, credits, or billing rules. A new-user allowance or promotional credits should be treated as a limited trial, not as permanent free access.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This is another reason to compare the whole workflow rather than one headline price. Resolution, duration, failed generations, and audio can change the real cost per usable clip. For the newer model family, the <a href=\"https:\/\/www.glbgpt.com\/hub\/wan-3-0-pricing\/\">Wan 3.0 pricing and API cost guide<\/a> provides more current planning context.<\/p>\n\n\n\n<h2 id=\"wan-2-5-open-source\" class=\"wp-block-heading\">Is Wan 2.5 Open Source?<\/h2>\n\n\n\n<p class=\"wan25 fact-check wp-block-paragraph\"><strong>No official Wan 2.5 model weights were listed in Alibaba&#8217;s public Wan repository when checked on September 3, 2026.<\/strong> \u30aa\u30d5\u30a3\u30b7\u30e3\u30eb <a href=\"https:\/\/github.com\/Wan-Video\/Wan2.2\">Wan2.2 GitHub repository<\/a> provides inference code, checkpoints, and an Apache 2.0 license for Wan 2.2 releases. It does not turn the separately hosted Wan 2.5 Preview endpoints into open-source downloads.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This is the common source of confusion behind the keyword &#8220;Wan 2.5 open source.&#8221; Wan is an open model family in a broad sense, but every numbered product is not automatically released under the same terms. API access means you can call a model. Open weights mean you can download and run its parameters. Those are different things.<\/p>\n\n\n\n<h2 id=\"wan-2-5-comfyui\" class=\"wp-block-heading\">Can You Run Wan 2.5 in ComfyUI?<\/h2>\n\n\n\n<p class=\"wan25 fact-check wp-block-paragraph\"><strong>There is no verified official local Wan 2.5 ComfyUI workflow without official Wan 2.5 weights.<\/strong> Alibaba&#8217;s Wan2.2 repository explicitly records ComfyUI integration for Wan2.2 text-to-video, image-to-video, text-image-to-video, speech-to-video, and animation tasks. That evidence applies to Wan 2.2, not Wan 2.5 Preview.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"623\" src=\"https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/03-wan22-open-source-comfyui-1024x623.webp\" alt=\"Official Wan2.2 repository noting ComfyUI integration\" class=\"wp-image-18894\" srcset=\"https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/03-wan22-open-source-comfyui-1024x623.webp 1024w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/03-wan22-open-source-comfyui-300x182.webp 300w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/03-wan22-open-source-comfyui-768x467.webp 768w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/03-wan22-open-source-comfyui-18x11.webp 18w, https:\/\/wp.glbgpt.com\/wp-content\/uploads\/2026\/09\/03-wan22-open-source-comfyui.webp 1100w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">A third-party ComfyUI node may still call a hosted Wan 2.5 API. That is an API wrapper, not local inference. Check where the node sends your prompt and source images, how it stores credentials, and how it charges for generations before installing it.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">If local control matters more than this exact version number, use an officially released Wan 2.2 checkpoint and its documented ComfyUI workflow. The trade-off is that the core Wan 2.2 text-to-video and image-to-video endpoints do not match Wan 2.5 Preview&#8217;s native synchronized-audio feature set.<\/p>\n\n\n\n<h2 id=\"wan-model-comparison\" class=\"wp-block-heading\">Wan 2.5 vs Newer Wan Models<\/h2>\n\n\n\n<figure class=\"wp-block-table wan25 table-wrap\"><table><thead><tr><th>\u30d0\u30fc\u30b8\u30e7\u30f3<\/th><th>\u3053\u308c\u3092\u4f7f\u3046\u6700\u5927\u306e\u7406\u7531<\/th><th>\u671f\u9593\u3068\u6c7a\u8b70<\/th><th>Local\/open workflow<\/th><\/tr><\/thead><tbody><tr><td>\u30ef\u30f3 2.2<\/td><td>Official checkpoints and ComfyUI support<\/td><td>Varies by checkpoint; core T2V\/I2V supports 480P or 720P<\/td><td>Yes, for officially released Wan 2.2 models<\/td><\/tr><tr><td>Wan 2.5 Preview<\/td><td>Short video with synchronized audio<\/td><td>5 or 10 seconds; 480P, 720P, 1080P<\/td><td>No official weights verified<\/td><\/tr><tr><td>\u30ef\u30f3 2.7<\/td><td>Longer hosted clips and multi-shot features<\/td><td>Up to 15 seconds; 720P or 1080P for T2V\/I2V<\/td><td>Hosted endpoints in Model Studio<\/td><\/tr><tr><td>Wan 3.0<\/td><td>Longer clips and broader reference inputs<\/td><td>2 to 30 seconds; 480P, 720P, 1080P<\/td><td>Hosted endpoints in Model Studio<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Wan 2.5 still makes sense when you need its exact hosted endpoint or want to reproduce an older workflow. For new projects, compare it with the latest option first. The <a href=\"https:\/\/www.glbgpt.com\/hub\/wan-3-0-vs-kling-3-0\/\">Wan 3.0 vs Kling 3.0 comparison<\/a> is a better starting point for a current buying decision.<\/p>\n\n\n\n<h2 id=\"wan-2-5-prompts\" class=\"wp-block-heading\">Wan 2.5 Prompt Checklist<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Use this order to keep the prompt readable:<\/p>\n\n\n\n<p class=\"wan25 prompt-formula wp-block-paragraph\"><code>Subject + action + setting + camera + lighting\/style + audio + constraints<\/code><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Example text-to-video prompt:<\/strong> A ceramic coffee cup sits on a dark wooden counter as steam curls upward. The camera makes a slow push-in. Warm morning light enters from the left. Quiet cafe ambience, a soft cup clink, no dialogue, no on-screen text.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Example image-to-video prompt:<\/strong> Keep the product shape and label unchanged. Condensation slowly forms on the bottle while the camera tracks from left to right. Cool studio lighting, shallow depth of field, subtle ice crackle and room tone. No extra objects and no text distortion.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Use one main action per short clip.<\/li>\n\n\n\n<li>Choose one camera move instead of combining several.<\/li>\n\n\n\n<li>Write exact audio cues rather than asking for &#8220;good sound.&#8221;<\/li>\n\n\n\n<li>Protect important product details with explicit constraints.<\/li>\n\n\n\n<li>Generate at a lower resolution while testing motion, then increase quality.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">For adult or restricted-content questions, separate model capability from platform policy. The <a href=\"https:\/\/www.glbgpt.com\/hub\/how-to-use-wan-generate-nsfw-videos\/\">Wan content-policy and prompt workflow guide<\/a> explains the practical limits without treating every generated request as acceptable.<\/p>\n\n\n\n<h2 id=\"wan-2-5-faq\" class=\"wp-block-heading\">\u3088\u304f\u3042\u308b\u8cea\u554f<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\">What is Wan 2.5?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Wan 2.5 is an Alibaba AI video model released as a hosted Preview service. Its official text-to-video and image-to-video endpoints create 5- or 10-second MP4 clips at up to 1080P and include synchronized audio generation.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Is Wan 2.5 an Alibaba AI video model?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Yes. Wan is Alibaba&#8217;s visual-generation model family, and Alibaba introduced Wan2.5 Preview at Apsara Conference 2025. The company presented four Wan2.5 models covering video generation, image generation, and image editing.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Is Wan 2.5 open source?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">No official Wan 2.5 weights were listed in Alibaba&#8217;s public Wan repository as of September 3, 2026. Open-source Wan 2.2 checkpoints exist, but that does not make the hosted Wan 2.5 Preview endpoints open source.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Can I use Wan 2.5 with ComfyUI?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Not as verified local inference without official Wan 2.5 weights. Some third-party nodes may call a hosted API, but that is different from running the model locally. Official ComfyUI integration is documented for released Wan 2.2 checkpoints.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Is Wan 2.5 free to use?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Wan 2.5 is not a permanently free downloadable model. Hosted access can use provider credits, trial quota, or paid billing. Check the live platform before generating because account eligibility, credit costs, and model availability can change.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Does Wan 2.5 support image-to-video generation?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Yes. The official <code>wan2.5-i2v-preview<\/code> endpoint turns an input image into a 5- or 10-second video. It supports 480P, 720P, and 1080P MP4 output at 30 fps with synchronized audio.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Should I still use Wan 2.5 in 2026?<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Use Wan 2.5 when you need that exact Preview endpoint or must reproduce an older workflow. For a new project, compare Wan 2.7 and Wan 3.0 first because Alibaba&#8217;s current catalog gives them longer duration or broader controls.<\/p>\n\n\n\n<h2 id=\"final-verdict\" class=\"wp-block-heading\">\u6700\u7d42\u8a55\u6c7a<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Wan 2.5 was an important step in Alibaba&#8217;s AI video push because it combined text-to-video and image-to-video generation with synchronized sound. Its official specifications remain useful: two hosted Preview endpoints, up to 1080P, 5- or 10-second duration, 30 fps, and MP4 output.<\/p>\n\n\n\n<script type=\"application\/ld+json\">{\n    \"@context\": \"https:\\\/\\\/schema.org\",\n    \"@type\": \"FAQPage\",\n    \"mainEntity\": [\n        {\n            \"@type\": \"Question\",\n            \"name\": \"What is Wan 2.5?\",\n            \"acceptedAnswer\": {\n                \"@type\": \"Answer\",\n                \"text\": \"Wan 2.5 is an Alibaba AI video model released as a hosted Preview service. Its official text-to-video and image-to-video endpoints create 5- or 10-second MP4 clips at up to 1080P and include synchronized audio generation.\"\n            }\n        },\n        {\n            \"@type\": \"Question\",\n            \"name\": \"Is Wan 2.5 an Alibaba AI video model?\",\n            \"acceptedAnswer\": {\n                \"@type\": \"Answer\",\n                \"text\": \"Yes. Wan is Alibaba's visual-generation model family, and Alibaba introduced Wan2.5 Preview at Apsara Conference 2025. The company presented four Wan2.5 models covering video generation, image generation, and image editing.\"\n            }\n        },\n        {\n            \"@type\": \"Question\",\n            \"name\": \"Is Wan 2.5 open source?\",\n            \"acceptedAnswer\": {\n                \"@type\": \"Answer\",\n                \"text\": \"No official Wan 2.5 weights were listed in Alibaba's public Wan repository as of September 3, 2026. Open-source Wan 2.2 checkpoints exist, but that does not make the hosted Wan 2.5 Preview endpoints open source.\"\n            }\n        },\n        {\n            \"@type\": \"Question\",\n            \"name\": \"Can I use Wan 2.5 with ComfyUI?\",\n            \"acceptedAnswer\": {\n                \"@type\": \"Answer\",\n                \"text\": \"Not as verified local inference without official Wan 2.5 weights. Some third-party nodes may call a hosted API, but that is different from running the model locally. Official ComfyUI integration is documented for released Wan 2.2 checkpoints.\"\n            }\n        },\n        {\n            \"@type\": \"Question\",\n            \"name\": \"Is Wan 2.5 free to use?\",\n            \"acceptedAnswer\": {\n                \"@type\": \"Answer\",\n                \"text\": \"Wan 2.5 is not a permanently free downloadable model. Hosted access can use provider credits, trial quota, or paid billing. Check the live platform before generating because account eligibility, credit costs, and model availability can change.\"\n            }\n        },\n        {\n            \"@type\": \"Question\",\n            \"name\": \"Does Wan 2.5 support image-to-video generation?\",\n            \"acceptedAnswer\": {\n                \"@type\": \"Answer\",\n                \"text\": \"Yes. The official wan2.5-i2v-preview endpoint turns an input image into a 5- or 10-second video. It supports 480P, 720P, and 1080P MP4 output at 30 fps with synchronized audio.\"\n            }\n        },\n        {\n            \"@type\": \"Question\",\n            \"name\": \"Should I still use Wan 2.5 in 2026?\",\n            \"acceptedAnswer\": {\n                \"@type\": \"Answer\",\n                \"text\": \"Use Wan 2.5 when you need that exact Preview endpoint or must reproduce an older workflow. For a new project, compare Wan 2.7 and Wan 3.0 first because Alibaba's current catalog gives them longer duration or broader controls.\"\n            }\n        }\n    ]\n}<\/script>\n<\/div>","protected":false},"excerpt":{"rendered":"<p>Quick answer: Wan 2.5 is Alibaba&#8217;s hosted AI video model for text-to-video and image-to-video generation. Its official Model Studio endpoints can produce 5- or 10-second MP4 clips at 480P, 720P, or 1080P, with synchronized audio generated alongside the visuals. The model remains in Alibaba Cloud&#8217;s documentation, although it is no longer the newest release in the Wan family. Much of the confusion around Wan 2.5 comes from its relationship to Alibaba&#8217;s open-source Wan models. They belong to the same family, but the official Wan 2.5 video endpoints were released as hosted Preview services. Alibaba has not published verified Wan 2.5 weights for local ComfyUI use, so access, licensing, and setup [&hellip;]<\/p>","protected":false},"author":17,"featured_media":18896,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_seopress_robots_primary_cat":"","_seopress_titles_title":"Wan 2.5: Alibaba's AI Video Generator Explained","_seopress_titles_desc":"Wan 2.5 explained with official specs, current access, image-to-video steps, and the real open-source and ComfyUI status. Check the facts before you use it.","_seopress_robots_index":"","footnotes":""},"categories":[7],"tags":[],"class_list":["post-18890","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-chat"],"_links":{"self":[{"href":"https:\/\/wp.glbgpt.com\/jp\/wp-json\/wp\/v2\/posts\/18890","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wp.glbgpt.com\/jp\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/wp.glbgpt.com\/jp\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/wp.glbgpt.com\/jp\/wp-json\/wp\/v2\/users\/17"}],"replies":[{"embeddable":true,"href":"https:\/\/wp.glbgpt.com\/jp\/wp-json\/wp\/v2\/comments?post=18890"}],"version-history":[{"count":1,"href":"https:\/\/wp.glbgpt.com\/jp\/wp-json\/wp\/v2\/posts\/18890\/revisions"}],"predecessor-version":[{"id":18895,"href":"https:\/\/wp.glbgpt.com\/jp\/wp-json\/wp\/v2\/posts\/18890\/revisions\/18895"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/wp.glbgpt.com\/jp\/wp-json\/wp\/v2\/media\/18896"}],"wp:attachment":[{"href":"https:\/\/wp.glbgpt.com\/jp\/wp-json\/wp\/v2\/media?parent=18890"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/wp.glbgpt.com\/jp\/wp-json\/wp\/v2\/categories?post=18890"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/wp.glbgpt.com\/jp\/wp-json\/wp\/v2\/tags?post=18890"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}