{"id":2028,"date":"2026-05-02T20:03:34","date_gmt":"2026-05-02T20:03:34","guid":{"rendered":"https:\/\/wiro.ai\/blog\/?p=2028"},"modified":"2026-09-27T17:01:36","modified_gmt":"2026-09-27T17:01:36","slug":"wan-2-2-fast-text-to-video-4-short-prompt-tests-480p","status":"publish","type":"post","link":"https:\/\/wiro.ai\/blog\/wan-2-2-fast-text-to-video-4-short-prompt-tests-480p\/","title":{"rendered":"Wan 2.2 Fast Text-to-Video: 4 Short Prompt Tests (480p)"},"content":{"rendered":"<h2>Wan 2.2 Fast Text-to-Video: 4 Short Prompt Tests (480p)<\/h2>\n<p>Wan 2.2 Fast text-to-video was tested here with four short 480p prompts aimed at different kinds of motion. The point was not to crown a winner from four clips. It was to make the trade-offs visible: a moving camera over water, an animal held in frame, liquid in close-up, and a stylized space with a pull-back shot. Those are common requests, and each can expose a different failure mode.<\/p>\n<nav><strong>In this post<\/strong><\/p>\n<ul>\n<li><a href=\"#setup\">Test setup<\/a><\/li>\n<li><a href=\"#results\">What the four outputs show<\/a><\/li>\n<li><a href=\"#reading\">How to read the results<\/a><\/li>\n<li><a href=\"#pick\">When to pick Wan 2.2 Fast<\/a><\/li>\n<\/ul>\n<\/nav>\n<h2 id=\"setup\">Test setup for Wan 2.2 Fast text-to-video<\/h2>\n<p>All four preserved outputs use <a href=\"https:\/\/wiro.ai\/models\/wan-ai\/wan2-2-ti2v-5b-text-to-video-fast\">Wan 2.2 Fast Text-to-Video on Wiro<\/a>. The original test fixed the resolution at 480p, the frame at 16:9, and the seed at 42. It used no negative prompt. Holding those choices steady makes the clips easier to compare, even though four prompts are not enough for a benchmark.<\/p>\n<ul>\n<li><strong>Model:<\/strong> <a href=\"https:\/\/wiro.ai\/models\/wan-ai\/wan2-2-ti2v-5b-text-to-video-fast\">Wan 2.2 Fast Text-to-Video<\/a><\/li>\n<li><strong>Resolution:<\/strong> 480p<\/li>\n<li><strong>Aspect ratio:<\/strong> 16:9<\/li>\n<li><strong>Seed:<\/strong> 42<\/li>\n<li><strong>Negative prompt:<\/strong> none<\/li>\n<\/ul>\n<p>The saved test record does not contain a per-output run time or a per-output Wiro cost. Neither figure is added here because it cannot be recovered honestly from the four embedded files. This post should therefore help with prompt and output selection, not with budgeting or speed comparisons. The Wan project&#8217;s <a href=\"https:\/\/github.com\/Wan-Video\/Wan2.2\" target=\"_blank\" rel=\"noopener\">official open-source repository<\/a> describes the wider Wan 2.2 family and its emphasis on motion and controllable visual style, but it does not turn these four clips into a measured performance test.<\/p>\n<h2 id=\"results\">What the four outputs show<\/h2>\n<h3>1. Stormy sea: slow forward camera move<\/h3>\n<figure><video controls preload=\"metadata\" src=\"https:\/\/wiro.ai\/blog\/wp-content\/uploads\/2026\/03\/wan-v1.mp4\"><\/video><figcaption>Output 1: stormy sea, tested as a slow forward camera move.<\/figcaption><\/figure>\n<p>The first output puts water, weather, and camera travel in the same request. It is useful because waves create repeating detail across most of the frame while a forward move asks the scene to keep changing perspective. Watch the horizon, foam bands, and distant cloud texture rather than only the center of the image. If those areas appear to slide as one flat layer, the shot will feel synthetic even when the water itself looks dramatic. For a practical prompt, the lesson is to name one camera direction and one subject of motion. Adding boats, people, birds, and multiple cuts would make this test harder to diagnose.<\/p>\n<h3>2. Tiger walk: subject tracking<\/h3>\n<figure><video controls preload=\"metadata\" src=\"https:\/\/wiro.ai\/blog\/wp-content\/uploads\/2026\/03\/wan-v2.mp4\"><\/video><figcaption>Output 2: tiger walk, used to inspect a centered moving subject.<\/figcaption><\/figure>\n<p>The second output changes the problem from environmental motion to a recognizable moving subject. A walking tiger tests whether the model can keep a body readable while the background changes and the shot follows it. Stripes, paws, tail position, and head shape are high-signal details. They make drift easier to spot than it would be on a plain-colored subject. This is the right kind of prompt for checking subject persistence: one animal, a clear action, and a stated framing goal. It is not a reliable test of complex interaction, crowd scenes, or exact anatomy under fast turns.<\/p>\n<h3>3. Macro orange juice pour: liquid and refraction<\/h3>\n<figure><video controls preload=\"metadata\" src=\"https:\/\/wiro.ai\/blog\/wp-content\/uploads\/2026\/03\/wan-v3.mp4\"><\/video><figcaption>Output 3: macro orange juice pour, focused on liquid detail and glass.<\/figcaption><\/figure>\n<p>The third output narrows the field of view. It asks for a close-up pour with splashes, bubbles, transparent glass, and reflections. That combination matters for product footage because the result can look convincing at a glance while breaking under close inspection. The useful places to inspect are the stream where it meets the glass, the edge of the liquid, and reflections that pass behind the action. Macro prompts benefit from restraint. Specify the liquid, container, lighting, and one motion. Avoid a long list of ingredients or props unless they are central to the shot.<\/p>\n<h3>4. Animated corn kernel room: stylized pull-back<\/h3>\n<figure><video controls preload=\"metadata\" src=\"https:\/\/wiro.ai\/blog\/wp-content\/uploads\/2026\/03\/wan-v4.mp4\"><\/video><figcaption>Output 4: animated corn kernel room, testing a stylized pull-back.<\/figcaption><\/figure>\n<p>The final output moves away from realism. It uses an animated room built around corn kernels, then asks the camera to pull back. Stylized prompts are valuable because they test visual consistency without depending on photorealistic skin, fur, or glass. The key question is whether the room still reads as one designed place as more of it enters the frame. Repeated shapes should feel intentional, and the color treatment should stay coherent. This kind of prompt suits playful explainers, short social clips, and concept work where a readable visual idea matters more than literal realism.<\/p>\n<h2 id=\"reading\">How to read the results<\/h2>\n<p>These four files cover four different pressure points. The sea clip tests broad texture and camera travel. The tiger clip tests a moving hero subject. The juice pour tests small-scale physical detail. The animated room tests style continuity as the camera reveals more space. A good result in one does not prove the same behavior in the others.<\/p>\n<table>\n<thead>\n<tr>\n<th>Output<\/th>\n<th>Primary check<\/th>\n<th>What to inspect<\/th>\n<th>Prompting takeaway<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>Stormy sea<\/td>\n<td>Camera travel<\/td>\n<td>Horizon, wave repetition, cloud texture<\/td>\n<td>State one camera move clearly.<\/td>\n<\/tr>\n<tr>\n<td>Tiger walk<\/td>\n<td>Subject continuity<\/td>\n<td>Paws, tail, stripes, head shape<\/td>\n<td>Keep one hero subject in frame.<\/td>\n<\/tr>\n<tr>\n<td>Orange juice pour<\/td>\n<td>Liquid detail<\/td>\n<td>Stream, splash edge, glass reflections<\/td>\n<td>Use a short macro setup.<\/td>\n<\/tr>\n<tr>\n<td>Corn kernel room<\/td>\n<td>Style continuity<\/td>\n<td>Repeated shapes, palette, room layout<\/td>\n<td>Name the visual idea and one move.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p>Prompt wording still matters more than adding adjectives. A compact instruction that identifies subject, action, scene, style, and camera move gives the generator a clear hierarchy. If an output matters, run a small seed sweep rather than assuming the first clip represents the model. Keep the prompt unchanged during that sweep, then change only one variable at a time.<\/p>\n<h2 id=\"pick\">When to pick Wan 2.2 Fast<\/h2>\n<p>Pick Wan 2.2 Fast when the job calls for a short 480p text-to-video draft with one readable action and a clear visual premise. It fits early creative exploration, storyboards, stylized motion ideas, and simple product or nature shots. The four tests suggest a sensible prompt pattern: one main subject, one main action, and one deliberate camera instruction.<\/p>\n<p>Choose a model or workflow built for higher resolution when final delivery needs more pixels. Choose a workflow with image guidance when an exact character, product, or starting composition must survive the clip. Choose a model with explicit audio features when synchronized speech or sound is the core requirement. Those are different constraints from the ones tested here, and they should drive the model choice before a prompt is written.<\/p>\n<p>For more text-to-video context, see <a href=\"https:\/\/wiro.ai\/blog\/wan-2-7-video-5-video-prompt-tests\/\">WAN 2.7 Video: 5 Video Prompt Tests<\/a>, <a href=\"https:\/\/wiro.ai\/blog\/ltx-video-vs-kling-vs-seedance\/\">LTX-Video vs Kling vs Seedance<\/a>, and <a href=\"https:\/\/wiro.ai\/blog\/top-5-text-to-video-apis-in-2026-1-prompt-each\/\">Top 5 Text-to-Video APIs in 2026<\/a>. To run the same kind of compact test, start with <a href=\"https:\/\/wiro.ai\/models\/wan-ai\/wan2-2-ti2v-5b-text-to-video-fast\">Wan 2.2 Fast Text-to-Video on Wiro<\/a>, keep the setup fixed, and judge each clip against the one behavior the prompt was designed to expose.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Wan 2.2 Fast Text-to-Video: 4 Short Prompt Tests (480p) Wan 2.2 Fast text-to-video was tested here with four short 480p prompts aimed&hellip;<\/p>\n","protected":false},"author":4,"featured_media":2041,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[52],"tags":[57,115],"class_list":["post-2028","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-model-reviews","tag-text-to-video","tag-wan2-2"],"_links":{"self":[{"href":"https:\/\/wiro.ai\/blog\/wp-json\/wp\/v2\/posts\/2028","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wiro.ai\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/wiro.ai\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/wiro.ai\/blog\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/wiro.ai\/blog\/wp-json\/wp\/v2\/comments?post=2028"}],"version-history":[{"count":2,"href":"https:\/\/wiro.ai\/blog\/wp-json\/wp\/v2\/posts\/2028\/revisions"}],"predecessor-version":[{"id":4178,"href":"https:\/\/wiro.ai\/blog\/wp-json\/wp\/v2\/posts\/2028\/revisions\/4178"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/wiro.ai\/blog\/wp-json\/wp\/v2\/media\/2041"}],"wp:attachment":[{"href":"https:\/\/wiro.ai\/blog\/wp-json\/wp\/v2\/media?parent=2028"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/wiro.ai\/blog\/wp-json\/wp\/v2\/categories?post=2028"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/wiro.ai\/blog\/wp-json\/wp\/v2\/tags?post=2028"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}