Qwen3.8-27B

Qwen@Alibaba_Qwen We promised open weights for Qwen3.8. Now, time to meet them! 🎉 ⚡

我们曾承诺过 Qwen3.8 将会开源权重。现在，是时候和大家见面了！🎉 ⚡

Qwen3.8-27B:

• A native multimodal dense model. With just 27B parameters, it outperforms Qwen3.7-Plus overall and shines in real-world coding & office workflows.
• 262K native context, easily extendable to 1M tokens via YaRN.
• Built for builders. Highly efficient, high-quality, and licensed under Apache 2.0. 🚀

Qwen3.8-27B：

• 一款原生多模态稠密模型。仅需 27B 参数，其整体性能超越了 Qwen3.7-Plus，并在实际编程和办公工作流中表现出色。
• 支持 262K 原生上下文，可通过 YaRN 轻松扩展至 1M token。
• 为开发者而生。高效、高质量，并采用 Apache 2.0 协议开源。🚀

The open weights for Qwen3.8-2.4T-A95B (Max-level) have also been released recently. Whether you’re shipping lightweight applications with Qwen3.8-27B locally or building agents with Qwen3.8-2.4T-A95B, they’re yours now! Download, deploy, and build something we haven’t imagined yet. 👀👇

Qwen3.8-2.4T-A95B（Max 级别）的开源权重也已于近期发布。无论你是想通过 Qwen3.8-27B 在本地部署轻量级应用，还是利用 Qwen3.8-2.4T-A95B 构建智能体，现在它们都属于你了！下载、部署，去创造我们尚未想象到的应用吧。👀👇

• Hugging Face： huggingface.co/collections/Qw…
• ModelScope： modelscope.cn/collections/Qw…

---

Unsloth AI@UnslothAI Thank you thank you Qwen team! Qwen3.8-27B is just amazing! 🥰 We made Dynamic GGUFs so you can run it locally: huggingface.co/unsloth/Qwen3.… The model is now also supported for running and fine-tuning in Unsloth Desktop.

感谢 Qwen 团队！Qwen3.8-27B 简直太棒了！🥰 我们制作了动态 GGUF 版本，方便你在本地运行：huggingface.co/unsloth/Qwen3.… 该模型现在也已支持在 Unsloth Desktop 中进行运行和微调。