# `Transmuter.EMLX`
[🔗](https://github.com/georgeguimaraes/transmuter/blob/v0.1.0/lib/transmuter/emlx.ex#L2)

Moves a model's params to EMLX and keeps 4-bit and 8-bit weights packed.

`MatMulNBits` weights become MLX quantized tensors. `Nx.dot` then runs
`mx::quantized_matmul` on the packed weights, so the float weights never
exist in memory. ONNX packs the lowest bits first, which is also the MLX
order, and an ONNX zero point becomes the MLX bias `-zero_point * scale`.

    model = Transmuter.load!({:hf, "onnx-community/Qwen3-0.6B-ONNX", "onnx/model_q4.onnx"}, backend: EMLX.Backend)

`Transmuter.load!/2` with `backend: EMLX.Backend` calls `params/2` for you.

Weights that MLX cannot hold (other bit widths, group sizes other than 32,
64 or 128, or a K that is not a whole number of groups) stay packed bytes
and are dequantized in the graph.

# `params`

Returns `model.params` on EMLX, with compatible `MatMulNBits` weights quantized.

## Options

  * `:device` - the EMLX device, `:gpu` (default) or `:cpu`

---

*Consult [api-reference.md](api-reference.md) for complete listing*