Qwen2.5 1.5B with GAI-Deployment-Toolkit-v2.0.8_qwen2.5-0.5b-1.5b-7b-v0.1.tar.gz &&neuropilot-sdk-basic-8.0.11-build20260211 BACKEND=“mdla5.3,edma3.6” L1_SIZE_KB=“256” NUM_MDLA=“1”
1: After Conversion and PTQ, Exec qwen2.5-1.5B_04_optional_inference.sh is OK
2:After Compilation &Inference On Device ,report the following error （note: the same environment, qwen2.5 0.5B & 7B is ok)
Using yaml config file: ../dla/config_qwen2.5_1. 5b_instruct
Current yaml config: ../dla/config_qwen2.5_1.5b.
Begin model init..FATAL: Check failed:
getExpectedNumInputs ( == this->getRun
nsure the model is correct. [jni/prebuilt/src/executor/1lm_
Aborted (core dumped)
instruct.yaml <<<<<<<<<<<
imeNumInputs Q (getExpectedNumInputs O=51, this->getRuntimel
xecutor.cpp:194]
S
lumInputs O=59). Number of inputs does not match, please e
lumInputs O=59). Number of inputs does not match
fixed .1.5B 3B also OK. it do’s need hellp .Thanks
Qwen2.5 1.5B with GAI-Deployment-Toolkit-v2.0.8_qwen2.5-0.5b-1.5b-7b-v0.1.tar.gz &&neuropilot-sdk-basic-8.0.11-build20260211 BACKEND=“mdla5.3,edma3.6” L1_SIZE_KB=“256” NUM_MDLA=“1”
1: After Conversion and PTQ, Exec qwen2.5-1.5B_04_optional_inference.sh is OK
2:After Compilation &Inference On Device ,report the following error （note: the same environment, qwen2.5 0.5B & 7B is ok)
Using yaml config file: ../dla/config_qwen2.5_1. 5b_instruct
Current yaml config: ../dla/config_qwen2.5_1.5b.
Begin model init..FATAL: Check failed:
getExpectedNumInputs ( == this->getRun
nsure the model is correct. [jni/prebuilt/src/executor/1lm_
Aborted (core dumped)
instruct.yaml <<<<<<<<<<<
imeNumInputs Q (getExpectedNumInputs O=51, this->getRuntimel
xecutor.cpp:194]
S
lumInputs O=59). Number of inputs does not match, please e
lumInputs O=59). Number of inputs does not match
