Skip to content

MNN:Feature: Support insert input scale for dynamic quant#4090

Open
jxt1234 wants to merge 1 commit intoalibaba:masterfrom
jxt1234:feature/insertquant
Open

MNN:Feature: Support insert input scale for dynamic quant#4090
jxt1234 wants to merge 1 commit intoalibaba:masterfrom
jxt1234:feature/insertquant

Conversation

@jxt1234
Copy link
Collaborator

@jxt1234 jxt1234 commented Dec 30, 2025

  1. 支持使用模型中的 qkv 量化参数(仅支持了 precision = normal 的情况)
  2. 支持使用模型中的 Convolution 输入量化参数

@jxt1234 jxt1234 force-pushed the feature/insertquant branch 2 times, most recently from 0903707 to 99b9e2c Compare January 4, 2026 12:48
MNN:Feature: support insert kv quant info for CPUAttention
@jxt1234 jxt1234 force-pushed the feature/insertquant branch from 99b9e2c to f412c03 Compare January 7, 2026 11:03
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant