Qxinyu
Qxinyu
Have you pushed QNN SDK libary to the device？And the broadcast error can be ignored.
You can modify line 81 in /source/backend/qnn/backend/QNNBackend.cpp from: if ((QNN_GET_ERROR_CODE(qnnInterface.logCreate(logCallback, QNN_LOG_LEVEL_ERROR, &logHandle)) != QNN_SUCCESS) || to： if ((QNN_GET_ERROR_CODE(qnnInterface.logCreate(logCallback, QNN_LOG_LEVEL_DEBUG, &logHandle)) != QNN_SUCCESS) || and then capture the logs during execution...
" DspTransport.openSession qnn_open failed, 0x80000406, prio 100" " IDspTransport: Unable to load lib 0x80000406" " DspTransport.getHandle failed, error 0x00000008" " createDspTransportInstance failed to config transport object" " queuesClose : SingleCoreSession...
Is the application still crashing now? We will test on Qualcomm platforms with Windows. On the Android platform, we're also using llm_demo for testing.
> Compute Shape Error for qnn/graph0.bin 请问下这个是为什么呢 你出错的模型是哪个呢，现在运行qwen3-vl的模型还有点问题，生成的模型会出现你这个问题，我们后续会更新转换工具。
> > [@Qxinyu](https://github.com/Qxinyu) Just an update, we have tried running it on android NPU (Snapdragon 8 Gen3), by following similar steps as in case of Windows and it worked perfectly....
方便提供模型给我们这里测试下吗
直接在线跑会有问题吗。
> 直接在线跑没问题。我只是想尝试一下静态模型和能否调用qnn的离线量化 > […](#) > > ---原始邮件--- > 发件人: ***@***.***> > 发送时间: 2025年12月10日(周三) 晚上8:00 > 收件人: ***@***.***>; > 抄送: ***@***.******@***.***>; > 主题: Re: [alibaba/MNN] 当前使用MNN2QNNModel转一个static静态模型，报错如下 (Issue [#4049](https://github.com/alibaba/MNN/issues/4049)) > > >...