ascend-memory-profiling git:20260910.55adacdA

Immutable. This exact content is served forever at /api/v1/blob/a77ef45604eac659.

---
name: "ascend-memory-profiling"
description: "Profile and attribute HBM memory usage on Ascend NPU for vLLM serving scenarios. Breaks down memory into fixed overhead, model weights, KV cache, HCCL buffers, activations, and runtime, with traceable evidence chains. Use for requests like \"分析显存占用\", \"显存 profiling\", \"HBM 用了多少\", \"内存各部分拆分\". Do not use for performance profiling (kernel timing, throughput), offline inference, or non-Ascend hardware."
---

<!-- Generated from .agents/skills/ascend-memory-profiling/SKILL.md. Do not edit. -->

# Ascend Memory Profiling

Read `.agents/skills/ascend-memory-profiling/SKILL.md` and only the references needed
for the current task. The canonical skill owns the workflow.