@shop-os/modelkit 0.4.1 → 0.4.2

package/dist/catalog.js

@@ -1367,6 +1367,7 @@ var OMNI_RATIOS = ["16:9", "9:16"];
 var GROK_RATIOS = ["1:1", "16:9", "9:16", "4:3", "3:4", "3:2", "2:3"];
 var H3_RATIOS = ["21:9", "16:9", "4:3", "1:1", "3:4", "9:16"];
 var H3_REFERENCE_RATIOS = ["adaptive", ...H3_RATIOS];
+var WAN_RATIOS = ["adaptive", "16:9", "4:3", "1:1", "3:4", "9:16"];
 var NEGATIVE_PROMPT = "blur, distort, and low quality";
 var FRAMES_EXCLUDE_REFERENCES = { framesExcludeReferences: true };
 var KLING_PROMPT = { referenceSyntax: "numbered", maxChars: 2500 };
@@ -1377,6 +1378,7 @@ var SEEDANCE_2_AUDIO = {
   audioMaxBytes: 15 * 1024 * 1024
 };
 var H3_PROMPT = { referenceSyntax: "numbered", maxChars: 5e4 };
+var WAN_PROMPT = { referenceSyntax: "numbered", maxChars: 2e4 };
 var OMNI_PROMPT = { referenceSyntax: "caption", maxChars: 5e4 };
 var BYTEDANCE_LOGO = "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/3bba5d81-4e57-4e71-911b-cdfa05acca2c_compressed.png";
 var GOOGLE_LOGO2 = "https://d31vvfpk2ppdyq.cloudfront.net/assets/image/192485d2-a76a-4657-9c06-930b2c70badd_compressed.png";
@@ -1982,6 +1984,48 @@ var VIDEO_MODELS = [
     // references instead of the adapter silently dropping the frame.
     capabilities: { ...FRAMES_EXCLUDE_REFERENCES }
   },
+  {
+    // One row for fal's three Wan 3.0 endpoints; the adapter picks text-, image- or
+    // reference-to-video from the inputs. Each takes aspect_ratio, so none is ignored.
+    // https://fal.ai/models/alibaba/wan-3.0/reference-to-video/api
+    id: "wan-3.0",
+    aliases: ["wan-3", "wan3"],
+    label: "Wan 3",
+    type: "video",
+    adapter: "fal",
+    providerModelId: "alibaba/wan-3.0",
+    lifecycle: "active",
+    exposure: ["cowork", "space_builder"],
+    constraints: {
+      aspectRatios: [...WAN_RATIOS],
+      resolutions: ["480p", "720p", "1080p"],
+      durations: { min: 2, max: 30 },
+      durationPresets: [5, 10, 15, 30],
+      references: {
+        images: 10,
+        videos: 5,
+        audios: 5,
+        videoSeconds: { total: 15 },
+        audioSeconds: { total: 15 }
+      },
+      supports: ["seed", "lastFrame", "generateAudio", "safetyChecker"],
+      prompt: WAN_PROMPT,
+      audio: "optional"
+    },
+    defaults: {
+      durationSeconds: 5,
+      resolution: "1080p",
+      aspectRatio: "adaptive",
+      generateAudio: true
+    },
+    usage: { ...VIDEO_USAGE },
+    pricingKind: "video_per_second",
+    display: {
+      provider: "alibaba",
+      description: "Alibaba Wan 3 video model for text-to-video, image-to-video (start/end frame), and reference-to-video from up to 10 images, 5 videos and 5 audio clips, with native audio; 2-30 seconds at up to 1080p."
+    },
+    capabilities: { ...FRAMES_EXCLUDE_REFERENCES, lastFrameRequiresFirstFrame: true }
+  },
   {
     // Duration 3-10s per trigger-workflows mapDuration; the endpoint takes no resolution.
     id: "kling-o1-v2v",