master

ascend-robot【msmodelslim】【feature】增加v3.2 w8a8-quarot最佳实践

698b0b78创建于 2025年12月25日历史提交

文件	最后提交记录	最后更新时间
deepseek-r1-distill	【msmodelslim】DeepSeek-R1-Distill-1.5B/7B 迁至V0一键量化 Co-authored-by: zhangz200102<zhangzheng183@huawei.com>	5 个月前
deepseek_v3	【bugfix】修复qwen3 moe旋转设置block_size报错问题 Co-authored-by: libarry<870390541@qq.com>	5 个月前
deepseek_v3_2	【msmodelslim】【feature】增加v3.2 w8a8-quarot最佳实践 Co-authored-by: zhangz200102<zhangzheng183@huawei.com>	5 个月前
default	【msmodelslim】【feature】【v1】 introduce quant service plugin subsystem	9 个月前
hunyuan_video	新增一键量化hunyuanvideo多模态模型适配器 Co-authored-by: m0_57943685<gaozihao3@huawei.com>	5 个月前
kimi_k2	Kimi-k2-instruct-0905-W8A8量化适配 Co-authored-by: Secluded_Ocean<tangchuxiao0709@qq.com>	5 个月前
qwen1_5	【msmodelslim】Qwen1.5-110B、Qwen2-7B、Qwen2-72B量化切换到V0一键量化 Co-authored-by: SuperMeixc<meixiaocun1@huawei.com>	5 个月前
qwen2	【msmodelslim】Qwen1.5-110B、Qwen2-7B、Qwen2-72B量化切换到V0一键量化 Co-authored-by: SuperMeixc<meixiaocun1@huawei.com>	5 个月前
qwen2_5	【msmodelslim】【feature】【v1】 introduce quant service plugin subsystem	9 个月前
qwen3	【msmodelslim】更新w8a8c8为pdmix+c8 Co-authored-by: libarry<870390541@qq.com>	7 个月前
qwen3_moe	【msmodelslim】【资料】新增qwen3-30B最佳实践 Co-authored-by: ylzzz<yelinzhong@huawei.com>	7 个月前
qwen3_next	【msmodelslim】适配qwen3-next离群值抑制 Co-authored-by: caishengcheng<caishengcheng@huawei.com>	5 个月前
qwen3_vl_moe	【msmodelslim】【feature】【part1】新增多模态理解模型一键量化服务和Qwen3-VL-MoE量化适配 Co-authored-by: xiaoheng181<eudemoniaxh@163.com>	5 个月前
qwq	【msModelSlim】【bugfix】一键量化资料修正 Co-authored-by: joejoezhou<zhourongchen1@huawei.com>	8 个月前
wan2_1	【msmodelslim】【bugfix】修复多模态生成模型Wan2.1资料或日志问题 Co-authored-by: xiaoheng181<eudemoniaxh@163.com>	8 个月前
wan2_2	【msmodelslim】【feature】Wan2.2 系列模型的 W8A8 量化 -- 检视意见修改 Co-authored-by: huyinghong 00566667<huyinghong1@huawei.com>	5 个月前
__init__.py	基础设施整理	9 个月前