# Unsloth Shrinks Qwen-Image-2.1-Turbo to 2.47 GB for Consumer GPUs — retained record

> Retained news record from news:ml-alphasignal.

Canonical: https://highsignal.app/data/records/d897a6aee8e922b3

[← sources](https://highsignal.app/data)


source: [news](https://highsignal.app/data/news) (news:ml-alphasignal)  published 2026-10-31 16:00 ingested 2026-11-01 02:56

[original source ↗](https://alphasignal.ai/news/unsloth-shrinks-qwen-image-2-1-turbo-to-2-47-gb-for-consumer-gpus)

## retained text



```
Unsloth packaged Qwen's 8-step distilled image model as GGUF quants, letting the 7B generator run from under 3GB up to full F16 on consumer hardware.
```



document fetched 2026-11-01 02:56