# `EMLXAxon.MLX4BitParams`
[🔗](https://github.com/elixir-nx/emlx/blob/v0.3.0/emlx_axon/lib/emlx_axon/mlx4bit_params.ex#L1)

Loads Qwen3 weights from an MLX-4bit safetensors checkpoint into Bumblebee
Axon params format (BF16, Bumblebee `{in, out}` key convention).

## How it works

MLX-4bit checkpoints store linear weights in `{out, in/8}` packed u32 format
(MLX's `{out, in}` convention, packing the in-dimension). Loading proceeds in
two steps:

1. Dequantize each weight to BF16 — yielding the logical shape `{out, in}`.
2. Transpose to `{in, out}` — Bumblebee's Axon convention where `Nx.dot/6`
   contracts the weight's first axis (in-features).

After calling `load/1`, pass the result through
`EMLXAxon.QuantizeParams.quantize/1` to re-apply 4-bit quantization in
Bumblebee's convention so `quantized_matmul` dispatch is active at serving
time.

## Usage

    {:ok, model_info} = Bumblebee.load_model({:local, mlx_path})
    params = EMLXAxon.MLX4BitParams.load(mlx_path)
    params = EMLXAxon.QuantizeParams.quantize(params)
    serving = Bumblebee.Text.generation(
      %{model_info | params: params}, tokenizer, gen_cfg, ...
    )

# `load`

```elixir
@spec load(Path.t()) :: Axon.ModelState.t()
```

Load Qwen3 params from an MLX-4bit checkpoint directory.

Returns `%Axon.ModelState{}` with BF16 tensors in Bumblebee key layout.
All linear kernels are transposed from MLX `{out, in}` to Bumblebee `{in, out}`.

---

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