# Unsloth releases Dynamic v3.0 quants for Qwen3.8-27B

_Wednesday, August 19, 2026 at 11:55 AM EDT · Products · Latest · Tier 2 — Notable_

![Unsloth releases Dynamic v3.0 quants for Qwen3.8-27B — Primary](https://unsloth.ai/docs/~gitbook/image?url=https%3A%2F%2F2815821428-files.gitbook.io%2F%7E%2Ffiles%2Fv0%2Fb%2Fgitbook-x-prod.appspot.com%2Fo%2Forganizations%252FHpyELzcNe0topgVLGCZY%252Fsites%252Fsite_mXXTe%252Fsocialpreview%252FOeFDVbzp4cgwaId7PbvL%252Funsloth%2520docs%2520pic.png%3Falt%3Dmedia%26token%3Df34a6bcc-db15-449a-bc2d-e0e8f52fe953&width=1200&height=630&sign=b1ca68fa&sv=2)

Unsloth released Dynamic v3.0 quantized versions of Qwen3.8-27B, replacing an earlier shared preview. The company said the GGUF files work with most inference engines, including llama.cpp and Unsloth Desktop, and use post-training quantization rather than quantization-aware training. Unsloth said its new calibration dataset, layer selection and quantization methods preserve more model quality at a given disk size. It also said smaller quantizations remove the MTP module to save about 500MB of disk space.

## Sources

- [unsloth.ai](https://unsloth.ai/docs/basics/dynamic-3.0-ggufs)

---
Canonical: https://techandbusiness.org/newswire/7k5A389xSqgKE7ZT7u0VLS
Retrieved: 2026-08-19T23:39:23.941Z
Publisher: Tech & Business (techandbusiness.org)