WebLLM model library Qwen3-4B-q4f16_1-ctx4k_cs1k-webgpu.wasm

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 An unknown or withdrawn model is refused with a 400 naming it; the app page then shows "downloads 5.8 MB · runs on your device" and /models.txt carries the attribution.

{
  "models": [
    {
      "id": "qwen3-4b-q4f16-1-ctx4k-cs1k-webgpu",
      "revision": "025bcaf3"
    }
  ]
}

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("qwen3-4b-q4f16-1-ctx4k-cs1k-webgpu@025bcaf3", "Qwen3-4B-q4f16_1-ctx4k_cs1k-webgpu.wasm", {
  onProgress: (p) => (bar.style.width = Math.round(p.ratio * 100) + "%"),
});                                                   // ArrayBuffer — SHA-256 verified, cached

// Later visits: await ss.models.status("qwen3-4b-q4f16-1-ctx4k-cs1k-webgpu@025bcaf3")  → { cached: [...], missing: [] }
// Background prefetch on a landing page: ss.models.warm("qwen3-4b-q4f16-1-ctx4k-cs1k-webgpu@025bcaf3")

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.

// web-llm fetches shards by URL and resumes by range; point its model record at the registry.
const appConfig = { model_list: [{
  model: "https://models.skillsafe.ai/qwen3-4b-q4f16-1-ctx4k-cs1k-webgpu@025bcaf3/",              // params_shard_*.bin + the wasm kernel come from here
  model_id: "qwen3-4b-q4f16-1-ctx4k-cs1k-webgpu",
  model_lib: "https://models.skillsafe.ai/qwen3-4b-q4f16-1-ctx4k-cs1k-webgpu@025bcaf3/Qwen3-4B-q4f16_1-ctx4k_cs1k-webgpu.wasm",
}] };
const engine = await webllm.CreateMLCEngine("qwen3-4b-q4f16-1-ctx4k-cs1k-webgpu", { appConfig, initProgressCallback: onProgress });

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/qwen3-4b-q4f16-1-ctx4k-cs1k-webgpu@025bcaf3/Qwen3-4B-q4f16_1-ctx4k_cs1k-webgpu.wasm

curl -sSL -o Qwen3-4B-q4f16_1-ctx4k_cs1k-webgpu.wasm "https://models.skillsafe.ai/qwen3-4b-q4f16-1-ctx4k-cs1k-webgpu@025bcaf3/Qwen3-4B-q4f16_1-ctx4k_cs1k-webgpu.wasm"
shasum -a 256 Qwen3-4B-q4f16_1-ctx4k_cs1k-webgpu.wasm   # 5ddf44e49b03e53e24fd29a45591850924346140452f60c29280190388571340