Repository navigation
[提议] 基于Qwen3.8-27B蒸馏新一代QwenPaw-Flash-9B #7254
Forestchenglin
started this conversation in
Ideas
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Qwen3.8-27B在工具调用、结构化输出和长上下文记忆上提升显著,但27B对消费级显卡多智能体部署仍不友好。当前QwenPaw-Flash-9B基于Qwen2.5,已难以匹配AgentScope最新能力。
建议基于Qwen3.8-27B蒸馏新一代Flash-9B,重点对齐:
原生工具调用与JSON Schema约束生成
AgentScope消息/服务格式预对齐
Q4_K_M量化下6.5GB显存稳定运行
32K+上下文多轮记忆保持
社区可协助提供AgentScope交互数据、基准测试及量化验证。请问团队是否有相关规划?是否开放社区协作蒸馏?
感谢考虑!
All reactions