Skip to content

[Tracking] PagedKVCache Quantization #2663 中的TVM版本 #2880

Closed
@XJY990705

Description

@XJY990705

Overview

我需要测试 #2663 中量化后的性能,从源码下载了https://github.com/davidpissarra/mlc-llm/tree/kv-cache-quantization 这个分支中制定的tvm版本
image
f5f048b版本,但是在编译mlc-llm时报错,应该是tvm的版本不匹配导致的。请问我应该怎样解决?

Action Items

  • [ ]

Links to Related Issues and PRs

https://github.com/davidpissarra/mlc-llm/tree/kv-cache-quantization

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    Status

    Done

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions