---
license: apache-2.0
language:
- multilingual
tags:
- translation
- quality-estimation
- reference-free
- comet
- cometkiwi
- pruning
base_model: Unbabel/wmt22-cometkiwi-da
pipeline_tag: translation
---

# wmt22-cometkiwi-da-int8

A compressed version of [Unbabel/wmt22-cometkiwi-da](https://huggingface.co/Unbabel/wmt22-cometkiwi-da) — a reference-free machine-translation quality estimation model (source + MT only, no human reference required).

**Lossless compression** — zero human-Pearson loss, ~40% smaller on disk via int8 alone.

## What's different from the base model

- ****No layer pruning** — all 24 XLM-R encoder layers retained. Compression comes entirely from dynamic int8 quantization + fp16 storage.**
- `layerwise_attention` rebuilt to mix only the surviving layers (embeddings + kept layer outputs).
- **Dynamic int8 quantization** on the XLM-R encoder + fp16 storage (cast back to fp32 at load before quant). No layer pruning — all 24 encoder layers retained.

## Accuracy

Benchmarked on 1200 stratified segments from [RicardoRei/wmt-da-human-evaluation](https://huggingface.co/datasets/RicardoRei/wmt-da-human-evaluation) (reference-free, src+mt only):

| Metric | This variant | Full cometkiwi |
|---|---|---|
| Pearson r vs human DA | **0.6404** | 0.6402 |
| Spearman vs human DA  | **0.6703** | 0.6698 |
| Pearson r vs full     | **0.9919** | 1.0000 |
| MAE vs full           | **0.0138** | 0.0000 |
| Params                | **565.1M** | 565.1M |
| On-disk size          | **~1130 MB** | ~2200 MB |

### All variants at a glance

| Variant | Pearson(human) | Pearson(full) | Size | When to use |
|---|---|---|---|---|
| [full base](https://huggingface.co/Unbabel/wmt22-cometkiwi-da) | 0.6402 | 1.0000 | ~2200 MB | reference quality |
| [`-int8`](https://huggingface.co/solailabs/wmt22-cometkiwi-da-int8) | **0.6404** | 0.9919 | ~1300 MB | **lossless compression** |
| [`-pruned-k2`](https://huggingface.co/solailabs/wmt22-cometkiwi-da-pruned-k2) | **0.6300** | 0.9784 | ~2100 MB | best-quality pruned |
| [`-pruned-k4`](https://huggingface.co/solailabs/wmt22-cometkiwi-da-pruned-k4) | 0.5642 | 0.8316 | ~2060 MB | aggressive prune |
| [`-pruned-k4-xs`](https://huggingface.co/solailabs/wmt22-cometkiwi-da-pruned-k4-xs) | 0.5544 | 0.8113 | ~1030 MB | smallest footprint |

## Usage

```python
# pip install "unbabel-comet" "setuptools<81" huggingface_hub
# export HF_TOKEN=<your_token>   # must have Unbabel/wmt22-cometkiwi-da access

from huggingface_hub import snapshot_download
import sys
folder = snapshot_download(repo_id="solailabs/wmt22-cometkiwi-da-int8")
sys.path.insert(0, folder)
from load import load_model

model = load_model(folder)
out = model.predict(
    [{{"src": "The meeting has been postponed until next week.",
       "mt":  "La réunion a été reportée à la semaine prochaine."}}],
    batch_size=8, gpus=0, progress_bar=False, num_workers=2,
)
print(out["scores"])
```

The loader re-downloads the base cometkiwi, drops the same encoder layers, optionally applies int8 dynamic quantization, then loads the weights shipped in this repo.

## Files

- `state_dict.pt` — pruned model weights
- `config.json`   — base model id, kept/dropped layer indices, quant flag, accuracy
- `load.py`       — drop-in loader
- `README.md`     — this file

## Gated base model

The base `Unbabel/wmt22-cometkiwi-da` is gated. You must accept its license on the Hub while logged in with the same account your `HF_TOKEN` belongs to — otherwise the base-model download inside `load.py` returns 403.

## Citation

**Base model:** [`Unbabel/wmt22-cometkiwi-da`](https://huggingface.co/Unbabel/wmt22-cometkiwi-da) by Unbabel.

```
@inproceedings{{rei-etal-2022-cometkiwi,
    title = "{{C}}omet{{K}}iwi: {{IST}}-{{U}}nbabel 2022 Submission for the Quality Estimation Shared Task",
    author = "Rei, Ricardo  and others",
    booktitle = "WMT 2022",
}}
```

Released under the same license as the base model (Apache 2.0).