Clone any viral video with AI agents. Not just a script, the whole workflow: swap the face, the words, the B-roll, ship 100 variants in one command, and get your 100M views.
TypeScript agentic-aiaiai-agentsai-videocompilerdslffmpeggenerative-aillmmarkup-languagemonorepoplugin-systemprogramming-languagetext-to-videotypescriptvideovideo-automationvideo-clonevideo-editingvideo-generation
docs(seedance): clarify 2.5 supports 1080p
rponed6a77d7ac839
diff --git a/docs/quickstart/generation.md b/docs/quickstart/generation.mdindex faa934d0..3c3c5ff4 100644--- a/docs/quickstart/generation.md+++ b/docs/quickstart/generation.md@@ -98,7 +98,7 @@ choose the exact model independently of three invocation shapes. All shapes consordinary `Text` prompt and output `{id.video}`.Seedance 2.5 uses the same Surfaces; it is not a Runtime substitution for another model. Its exact-contract accepts 480p/720p, up to 30 image, 10 video and 10 audio references, and either `-1` for+contract accepts 480p/720p/1080p, up to 30 image, 10 video and 10 audio references, and either `-1` formodel-selected duration or an integer from 4 through 30 seconds:```svmldiff --git a/docs/zh/quickstart/generation.md b/docs/zh/quickstart/generation.mdindex a20af354..ec16c1b4 100644--- a/docs/zh/quickstart/generation.md+++ b/docs/zh/quickstart/generation.md@@ -87,7 +87,7 @@ hypit measure main.svml --segment hook --language en --pace normal --rounding roSeedance 只暴露模型能力,不暴露“口播”“B-roll”等创作用途。`standard`、`fast`、`mini`和 `2.5` 选择精确模型;调用形状则独立分为三种。三者都消费完整的普通 `Text` Prompt,并输出 `{id.video}`。-Seedance 2.5 复用同样的 Surface,而不是由 Runtime 把别的模型偷偷替换成 2.5。它的精确合同支持 480p/720p,最多 30 张参考图、10 段参考视频、10 段参考音频;时长可写 `-1`+Seedance 2.5 复用同样的 Surface,而不是由 Runtime 把别的模型偷偷替换成 2.5。它的精确合同支持 480p/720p/1080p,最多 30 张参考图、10 段参考视频、10 段参考音频;时长可写 `-1`交给模型选择,也可明确写 4–30 秒的整数:```svmldiff --git a/packages/provider-hypihub/README.md b/packages/provider-hypihub/README.mdindex f3f56c98..23b0538e 100644--- a/packages/provider-hypihub/README.md+++ b/packages/provider-hypihub/README.md@@ -11,6 +11,10 @@ admits them into the current Build's working byte area. Image references use Hyp`reference_image_urls`, `reference_videos`, and `reference_audios` fields (with `ref_video_url`for one video). First/last-frame images use `first_frame` and `last_frame`.+Seedance 2.5 (`@hypit/seedance` model `2.5`) maps to `bytedance/seedance-2-5` and supports+`480p`, `720p` and `1080p`. The Provider passes the authored `resolution` to `POST /v1/videos`;+omitting it in the Seedance Surface defaults to `720p`.+The current HypiHub GPT Image 2 route exposes the same request surface as its KIE upstream:| Resolution | Ratios unavailable at this Endpoint | `background` |diff --git a/packages/provider-hypihub/test/provider.test.ts b/packages/provider-hypihub/test/provider.test.tsindex 126b6fae..32b62c39 100644--- a/packages/provider-hypihub/test/provider.test.ts+++ b/packages/provider-hypihub/test/provider.test.ts@@ -34,6 +34,37 @@ async function endpointFor(request: Need, fetch: typeof globalThis.fetch): Promireturn resolution.registration.endpoint;}+test("HypiHub Seedance 2.5 resolves and submits 1080p without downgrading", async () => {+ const request: Need = {+ ...need(sealSeedanceRequest("seedance-2.5", {+ prompt: ["A presenter speaks to camera."], resolution: ["1080p"], aspectRatio: ["9:16"],+ duration: [5], generateAudio: [true], webSearch: [false],+ }) as unknown as CanonicalValue),+ capability: seedanceEndpoints.v25!.capability,+ returns: seedanceEndpoints.v25!.returns,+ };+ let submitted: unknown;+ const endpoint = await endpointFor(request, async (input, init) => {+ const url = String(input);+ if (url.endsWith("/models/bytedance%2Fseedance-2-5")) return Response.json({ endpoints: ["videos"] });+ if (url.endsWith("/videos")) {+ submitted = JSON.parse(String(init?.body));+ return Response.json({ id: "job_seedance25_1080p", status: "queued" });+ }+ throw new Error(`Unexpected request ${url}`);+ });+ const started = await endpoint.start({+ command: { kind: "fulfill-need", id: "command:seedance25", need: request },+ need: request, resources: new MemoryResourceStore(),+ credentials: { apiKey: { secret: "test-key" } }, operation: "operation:seedance25",+ });+ assert.equal(started.status, "pending");+ assert.deepEqual(submitted, {+ model: "bytedance/seedance-2-5", prompt: "A presenter speaks to camera.",+ resolution: "1080p", aspect_ratio: "9:16", seconds: 5, generate_audio: true, web_search: false,+ });+});+test("HypiHub portrait matting uses the video job lifecycle and stores transparent output", async () => {const resources = new MemoryResourceStore();const source = await resources.put(new Uint8Array([1, 2, 3]), "video/mp4");diff --git a/packages/seedance/test/surface.test.ts b/packages/seedance/test/surface.test.tsindex dcbff663..2adaf0be 100644--- a/packages/seedance/test/surface.test.ts+++ b/packages/seedance/test/surface.test.ts@@ -47,9 +47,9 @@ test("TextVideo exposes prompt-only generation without inventing a usage", asyncassert.equal(Object.keys(result.components[0]?.inputs ?? {}).some((name) => name.includes("media")), false);});-test("Seedance 2.5 is an exact model with auto or 4-30 second duration", async () => {+test("Seedance 2.5 is an exact model with 1080p and auto or 4-30 second duration", async () => {const automatic = await decode(- '<seedance:TextVideo id="motion" model="2.5" prompt={direction} duration="-1" resolution="720p" aspect-ratio="9:16"/>',+ '<seedance:TextVideo id="motion" model="2.5" prompt={direction} duration="-1" resolution="1080p" aspect-ratio="9:16"/>',decodeSeedanceTextVideoSurface,);assert.deepEqual(automatic.components[0]?.outputs, { video: "motion.video" });