源神歸位!阿裏開源 Qwen3.8-27B,24GB 顯卡可跑
三大亮點:
1️⃣ 一張消費級顯卡就能跑
270 億參數,4-bit 量化後約 15—18GB,RTX 3090/4090 等 24GB 顯卡即可本地部署。
代碼、合同和內部數據不用上傳雲端。
實際速度取決於量化格式、上下文及推理框架,「40 tokens/s」並非所有配置都能達到。
2️⃣ 體積更小,能力密度更高
原生支持文本、圖片和視頻,擁有 262K 上下文。
在編程、辦公和電腦操作等測試中,多項成績超過體量更大的 Qwen3.7-Plus。
3️⃣ 免費開源,還能商用
采用 Apache 2.0 協議,可修改、私有化部署並用於商業產品。
Pinterest 此前定製 Qwen3-VL,部分 AI 工作負載成本降低約 90%,證明開源模型真正的價值,是企業能夠按需裁剪並掌握控製權。
過去只能向巨頭租用智能,現在普通人也能真正擁有智能。
一張顯卡跑旗艦,源神,歸位。
#
AI# #
Qwen# #
WEEX#
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:
- ModelScope:
顯示更多