Runs in the browser · Embeddings
Run GTE Multilingual Base in your browser
GTE Multilingual Base downloads 443.7 MB over WebGPU (q4f16), or 324.6 MB over WebAssembly (uint8), to turn text into vectors for search and comparison entirely in the tab with Transformers.js. No install, no server.
Embeddings · onnx-community/gte-multilingual-base
- Parameters
- 305M
- Pipeline task
- feature-extraction
Reading
Param count is the model card's stated figure (no safetensors metadata to read directly), so treat it as approximate.
The WASM build downloads smaller here: 324.6 MB (uint8) against 443.7 MB (q4f16) for WebGPU. The CPU-fallback quant compresses tighter than the GPU pick for this model.
1.45 MB/M params against a 1.03 MB/M embeddings median: dense for its parameter count.
The quant ladder spans 3.7x: 324.6 MB (uint8) to 1.17 GB (fp32).
Will it run in your browser?
Checking for WebGPU support…
All measured variants
| Variant | Size |
|---|---|
| uint8 WASM pick | 324.6 MB |
| q4f16 WebGPU pick | 443.7 MB |
| fp16 | 598.9 MB |
| q8 | 649.1 MB |
| bnb4 | 826.1 MB |
| q4 | 832.8 MB |
| fp32 | 1.17 GB |
Sizes measured from the HuggingFace API file tree for onnx-community/gte-multilingual-base, not estimated.
Use it with Transformers.js
import { pipeline } from "@huggingface/transformers";
const pipe = await pipeline("feature-extraction", "onnx-community/gte-multilingual-base", {
device: "webgpu", // usually falls back to "wasm"; wrap in try/catch for production
dtype: "q4f16", // use "uint8" for the WASM build
});
const result = await pipe(/* your input */);
Requires npm install @huggingface/transformers (or the CDN build).
Sources
Same job, different size
FAQ
WebGPU or WASM for GTE Multilingual Base, in practice?
Chrome, Edge, and Safari 26+ run the WebGPU build (q4f16) on the GPU. Firefox has WebGPU on Windows and Apple Silicon Macs but not everywhere yet. Any browser without a WebGPU adapter falls back automatically to the WebAssembly build (uint8) on CPU: same model, slower to load, slower to run.
What does q4f16 mean for GTE Multilingual Base?
The WebGPU build here uses q4f16: 4-bit weights with some layers kept at 16-bit for stability: WebGPU's usual smallest clean build. The WASM fallback uses uint8: 8-bit integers: about a quarter the size of fp32, with a smaller quality trade than 4-bit.
Sizes measured 2026-08-01 from the HuggingFace API. Last validated 2026-08-03. See all browser models or the methodology.