U2-Net-p background removal (rembg)

Add to the body of POST /v1/apps/{slug}/releases (or a release session). An unknown or withdrawn model is refused with a 400 naming it; the app page then shows "downloads 4.4 MB · runs on your device" and /models.txt carries the attribution.

{
  "models": [
    {
      "id": "u2netp",
      "revision": "cd3a3d67"
    }
  ]
}

Streams with progress, verifies the SHA-256 against the catalogue, keeps a durable copy in the app's Cache API and reports whether the bytes came from cache or the network.

<script src="/sdk.js"></script>  <!-- vendored from https://skillsafe.ai/apps-sdk/v1.js -->

const ss = SkillSafe.init({ slug: "your-app" });
const caps = await ss.models.capabilities();        // { webgpu, wasm, wasmSimd, storage }
const bytes = await ss.models.load("u2netp@cd3a3d67", "u2netp.onnx", {
  onProgress: (p) => (bar.style.width = Math.round(p.ratio * 100) + "%"),
});                                                   // ArrayBuffer — SHA-256 verified, cached

// Later visits: await ss.models.status("u2netp@cd3a3d67")  → { cached: [...], missing: [] }
// Background prefetch on a landing page: ss.models.warm("u2netp@cd3a3d67")

The runtime itself ships in your bundle as a vetted {path, sha256} reference — only the weights come from the registry. Pass the bytes in; do not re-fetch by URL.

// onnxruntime-web 1.17.0 ships in your bundle as a vetted {path, sha256} reference —
// dist/ort.wasm.bundle.min.mjs + dist/ort-wasm-simd-threaded.wasm (or the .jsep pair for WebGPU).
import * as ort from "./ort.wasm.bundle.min.mjs";
ort.env.wasm.wasmPaths = { wasm: "/ort-wasm-simd-threaded.wasm" };  // object form, never a prefix string
ort.env.wasm.numThreads = 1;                                        // app hosts have no COOP/COEP

const bytes = await ss.models.load("u2netp@cd3a3d67", "u2netp.onnx", { onProgress });
const session = await ort.InferenceSession.create(new Uint8Array(bytes), {
  executionProviders: ["wasm", "webgpu"],
});

// Feeds are named after the graph inputs (read from the file, see the signature above):
const feeds = {
  "input.1": new ort.Tensor("float32", new Float32Array(n), [1, 3, 320, 320]),
};
const out = await session.run(feeds);
const 1959 = out["1959"].data;  // float32 [1, 1, 320, 320]
const 1960 = out["1960"].data;  // float32 [1, 1, 320, 320]
const 1961 = out["1961"].data;  // float32 [Sigmoid1961_dim_0, Sigmoid1961_dim_1, Sigmoid1961_dim_2, Sigmoid1961_dim_3]
const 1962 = out["1962"].data;  // float32 [Sigmoid1962_dim_0, Sigmoid1962_dim_1, Sigmoid1962_dim_2, Sigmoid1962_dim_3]
const 1963 = out["1963"].data;  // float32 [Sigmoid1963_dim_0, Sigmoid1963_dim_1, Sigmoid1963_dim_2, Sigmoid1963_dim_3]
const 1964 = out["1964"].data;  // float32 [Sigmoid1964_dim_0, Sigmoid1964_dim_1, Sigmoid1964_dim_2, Sigmoid1964_dim_3]
const 1965 = out["1965"].data;  // float32 [Sigmoid1965_dim_0, Sigmoid1965_dim_1, Sigmoid1965_dim_2, Sigmoid1965_dim_3]

Immutable, credential-free, Access-Control-Allow-Origin: *, range requests honoured. The host refuses requests carrying a foreign Origin; a plain curl is fine.

https://models.skillsafe.ai/u2netp@cd3a3d67/u2netp.onnx

curl -sSL -o u2netp.onnx "https://models.skillsafe.ai/u2netp@cd3a3d67/u2netp.onnx"
shasum -a 256 u2netp.onnx   # 309c8469258dda742793dce0ebea8e6dd393174f89934733ecc8b14c76f4ddd8

The Hugging Face repo these bytes were vetted from, pinned at cd3a3d6767a7. Use it directly for your own site or notebook. A SkillSafe app must load from the registry: its CSP allows models.skillsafe.ai only, and that copy is what the SDK verifies, the browser shares across apps and the app page discloses. SkillSafe keeps hosting its own copies and the API either way.

// Outside SkillSafe apps only: app hosts allow connect-src models.skillsafe.ai, not huggingface.co.
// Same bytes as the registry copy — the SHA-256 on the Hugging Face file page is 309c8469258dda74…
const res = await fetch("https://huggingface.co/tomjackson2023/rembg/resolve/cd3a3d6767a7859efea31ef0f2f373582cf06d82/u2netp.onnx");
const bytes = new Uint8Array(await res.arrayBuffer());
const session = await ort.InferenceSession.create(bytes, { executionProviders: ["wasm", "webgpu"] });

https://huggingface.co/tomjackson2023/rembg/resolve/cd3a3d6767a7859efea31ef0f2f373582cf06d82/u2netp.onnx