Skip to main content

model-compress-skill

太极平台「模型压缩」子 skill —— 通过 MCP 协议对模型进行数值精度压缩(量化,如 W4A8-FP8 / W8A8-FP8 等 W{n}A{n}-{精度} 策略)。当用户提及"模型压缩 / 量化 / W4A8 / W8A8 / FP8 / GPTQ / AWQ / SmoothQuant / INT8 / INT4 / 知识蒸馏 / pruning / quantization / distillation"等关键词时,应使用本 skill。

Jump to install

Source facts

Repository
liuhanzuo/Mixture-of-Memory
Last source activity
August 17, 2026 at 04:21
Detected SKILL.md language
Chinese
Stars
4
Forks
1

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.