---
license: apache-2.0
base_model: Qwen/Qwen3.5-122B-A10B
library_name: transformers
tags:
- dashq
- quantized
- post-training-quantization
---

# Qwen3.5-122B-A10B-DASHQ-INT2-g32

This repository contains a DASH-Q packed quantized checkpoint for `Qwen/Qwen3.5-122B-A10B`.

DASH-Q checkpoints require the lightweight DASH-Q runtime package for loading. They are not plain Transformers checkpoints because linear layers are stored as `PackedQuantizedLinear` modules.

## Install

```bash
pip install git+https://github.com/JaeminK/dashq.git
```

## Load

```python
from dashq import load_quantized

model, tokenizer = load_quantized(
    "jkim96/Qwen3.5-122B-A10B-DASHQ-INT2-g32",
    device_map="auto",
)
```

## Quantization

| Field | Value |
| --- | --- |
| Base model | `Qwen/Qwen3.5-122B-A10B` |
| Bits | `2` |
| Group size | `32` |
| Scale/zero dtype | `float16` |
| Calibration dataset | `wikitext2` |
| Calibration samples | `128` |
| Sequence length | `2048` |
| Original size | `250.1733 GB` |
| Quantized size | `49.2493 GB` |

## Evaluation

| Metric | Value |
| --- | ---: |
| `wikitext2_ppl` | 6.8874 |
| `zero-shot accuracy avg` | 69.0405 |
| `arc_challenge` | 62.4573 |
| `arc_easy` | 83.9646 |
| `commonsense_qa` | 71.2531 |
| `hellaswag` | 80.8006 |
| `lambada_openai` | 71.4341 |
| `openbookqa` | 45.4000 |
| `piqa` | 80.8487 |
| `truthfulqa_mc2` | 51.7249 |
| `winogrande` | 73.4807 |