# Qwen releases FP8 package for 27B vision-language model

_Published Sunday, August 16, 2026 at 7:31 AM EDT · AI, Products · Latest · Tier 2 — Notable_

![Qwen releases FP8 package for 27B vision-language model — Primary](https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Qwen/Qwen3.8-27B-FP8.png)

A Qwen repository provides FP8-quantized weights and configuration files for Qwen3.8-27B in the Hugging Face Transformers format. The repository says the 27-billion-parameter vision-language model is compatible with Transformers, vLLM, SGLang and TokenSpeed, and has a native context length of 262,144 tokens that can be extended to 1 million tokens.

It says the weights use fine-grained FP8 quantization with a block size of 128. The repository also says a hosted Qwen Cloud version with a 1-million-token default context and built-in tools is planned, but that service is coming soon.

## Sources

- [huggingface.co](https://huggingface.co/Qwen/Qwen3.8-27B-FP8)
- [Techmeme](https://www.techmeme.com/260817/p4#a260817p4)

---
Canonical: https://techandbusiness.org/newswire/79pqeCC86JQaYQMOvDAQ4I
Published: 2026-08-16T11:31:51.509Z
Story chronology: 2026-08-13T17:35:09.000Z
Retrieved: 2026-09-30T16:27:46.068Z
Publisher: Tech & Business (techandbusiness.org)