بچه‌های تیم Qwen علی‌بابا سر قولشون موندن و وزن‌های مدل چندوجهی Qwen3.8-27B رو همراه با مدل غول‌پیکر Qwen3.8-2.4T-A95B به صورت کاملاً متن‌باز و با لایسنس آپاچی ۲.۰ منتشر کردن

We promised open weights for Qwen3.8. Now, time to meet them! 🎉 ⚡ 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. 🚀 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. 👀👇 - Hugging Face： https://huggingface.co/collections/Qwen/qwen38 - ModelScope： https://www.modelscope.cn/collections/Qwen/Qwen38