---
title: "ClikaRT::quant"
sidebar_label: "quant"
description: "Namespace ClikaRT::quant"
---

<!-- Generated by tools/doxygen2mdx from Doxygen XML. Do not edit. -->

<span class="api-kind api-kind--namespace">namespace</span>

The quantized-checkpoint import taxonomy: the parsed `quantization_config` facts and the routing enums the split factories consume.

## Classes

| Name | Description |
| --- | --- |
| [`QuantizationConfig`](./QuantizationConfig.md) | The checkpoint-level facts a HF `quantization_config` dict carries: the factory arguments the dict OWNS, normalized across the exporters' spellings. Per-weight facts (the tensors themselves, a bitsandbytes `quant_state` blocksize, a GPTQ weight's input extent) stay with the state dict; this struct is the dict side only. |

## Enumerations

### enum QuantizedCheckpointMethod {#QuantizedCheckpointMethod}

<div class="api-signature">

`enum class` **`QuantizedCheckpointMethod`**

</div>

Which serving import route a parsed `quantization_config` resolves to; one enumerator per split factory above.

| Enumerator | Description |
| --- | --- |
| `Gptq` | [`make_quantized_gptq`](../../index.md#make_quantized_gptq) (per weight: qweight/qzeros/scales\[/g\_idx\]) |
| `Awq` | [`make_quantized_awq`](../../index.md#make_quantized_awq) (the AutoAWQ GEMM layout) |
| `Bnb4` | [`make_quantized_bnb4`](../../index.md#make_quantized_bnb4) (bitsandbytes nf4/fp4) |

<div class="api-declared">

Declared in `ClikaRT/compute/q_tensor.h`, line 291

</div>

## Functions

### parse\_quantization\_config() {#parse_quantization_config}

<div class="api-signature">

[`QuantizationConfig`](./QuantizationConfig.md) **`parse_quantization_config`**(`std::string_view` `json_text`)

</div>

<div class="api-declared">

Declared in `ClikaRT/compute/q_tensor.h`, line 377

</div>

## `ClikaRT/compute/q_tensor.h` {#q_tensor-h}

```cpp
#include <ClikaRT/compute/q_tensor.h>
```

`QTensor`: the typed public view of a quantized weight. A quantized checkpoint entry (e.g. a block-quantized tensor from `io::load_gguf`) rides a plain `Tensor` whose element data is the packed byte stream; `QTensor` names what that payload IS: the packed data plus the quantization scheme and the logical element shape it encodes. Build one with `quantized_view(tensor)` and hand it to the primitives that consume quantized weights (`QLinearWoQ`).

── ORIENTATION MAP: read this before wiring a new checkpoint format ──────

Each ingestion entry below takes its checkpoint format's NATIVE orientation; that is what makes adoption zero-copy (a unified convention would force a transpose of somebody's checkpoint bytes), so the orientations deliberately DIFFER across entries, and mirroring the nearest sibling is how a new format gets wired wrong. The map, for the entries that take a shape argument (the ones a caller can mis-orient):

| entry | 2-D shape argument reads as |
| --- | --- |
| make\_quantized | \{in\_features, out\_features\} (k-major file |
| make\_quantized\_mxfp4 | order; row-contiguous dim FIRST) |

make\_quantized\_nvfp4 \| make\_quantized\_fp8 \| \{in\_features, out\_features\}; a checkpoint \| storing \[out, in\] transposes BEFORE entry make\_quantized\_fp8\_blocked \| \{out\_features, in\_features\}; N-MAJOR, the \| checkpoint's own order (blocks run along \| in\_features, the trailing axis) make\_quantized\_bnb4 \| the checkpoint's own \[out, in\] make\_quantized\_affine \| \{out\_features, in\_features\}; N-MAJOR, the \| orientation this scheme's checkpoints ship

A stacked expert weight keeps its entry's file order with the expert dim LAST (`{K, N_per_expert, E}`). The word-split adopts (`make_quantized_gptq` / `_awq` / `_ct_pack`) take no shape argument; the stored word layout IS their contract, and each entry's own doc names the adopted orientation.

THREE AUTHORITIES, one weight: (1) the ingestion entry above fixes how the payload's bytes are LABELED; (2) the module bind (`nn::QLinearWoQ` / the linear family) states the layout it expects of the labeled weight; (3) the runtime canonicalizes orientation ONCE at first use (the `transpose_b` fold), or, for a scheme whose canonical layout is n-major, keeps the bound layout as the served form, so a mislabeled weight is refused by the structural gates (loudly, at load), never multiplied in the wrong orientation. When adding a format, pick the entry whose orientation matches the CHECKPOINT's bytes; never re-shape the bytes to fit a different entry.