{"slug":"qwen3-8","name":"Qwen3.8","nameZh":"Qwen3.8（通义千问开放模型：27B 稠密 / Flash-Next）","category":"models","categoryName":"Open-weight models","audience":["ai-era"],"primary":"ai-era","openClosed":"open","pricing":"open-source","license":"Apache-2.0（27B）；Flash-Next 为 Qwen Community License","maintained":"yes","officialUrl":"https://huggingface.co/Qwen/Qwen3.8-27B","repo":"https://github.com/QwenLM/Qwen3.8","latestVersion":null,"latestRelease":null,"stars":4309,"starsAsOf":"2026-10-22","description":"The latest generation of Alibaba’s open Qwen models (27B posted on HF 2026-08-05, Flash-Next 2026-08-24): Qwen3.5-style hybrid Gated DeltaNet + gated attention with multi-token prediction, native image/video understanding, thinking on by default (switchable per request, depth via reasoning_effort), 262K native context extendable to 1M. The 27B dense model is Apache-2.0 and runs on Transformers, vLLM and SGLang; one of the most-liked models on HF (18k+)","descriptionZh":"阿里通义千问目前最新的开放模型代（HF 上 27B 于 2026-08-05 发布，Flash-Next 于 08-24 发布）：基于 Qwen3.5 架构（Gated DeltaNet 线性注意力 + 门控注意力混合、多 token 预测），原生图像/视频理解，思考模式默认开、可按请求关闭或用 reasoning_effort 调深度；原生 262K 上下文，可扩到 1M。27B 稠密版 Apache-2.0，兼容 Transformers、vLLM、SGLang；是 HF 本周点赞最多的模型之一（18k+）","callMethods":[],"platforms":[],"addedAt":"2026-10-22","updatedAt":"2026-10-22","topics":[],"language":null,"lastPush":"2026-08-17","formerly":null,"packages":[],"recentVersions":[],"alternatives":["deepseek-v3","qwen3","gpt-oss","qwen3-coder","kimi-k2","llama-models","ultralytics","gemma"],"stacks":[],"scenarios":[],"problems":[],"reasonZh":null,"sources":["editorial","github-pool"],"record":"https://aimap.page/api/v1/software/qwen3-8.json","pages":{"en":"https://aimap.page/en/software/qwen3-8/","zh-hans":"https://aimap.page/zh-hans/software/qwen3-8/"}}