[QNN] Using QWEN3-1.7B model inference on Qualcomm 8550, the correct answer cannot be generated.

[jincan2010]

🐛 Describe the bug

model conversion
python examples/qualcomm/oss_scripts/llama/llama.py -b build-android -m SM8550 --temperature 0 --model_mode hybrid --prefill_ar_len 128 --max_seq_len 1024 --decoder_model qwen3-1_7b --prompt "I would like to learn python, could you teach me with a simple example?" --tasks wikitext --limit 1

model inferrence
python examples/qualcomm/oss_scripts/llama/llama.py -b build-android -s 117.149.248.154:12146 -m SM8550 --model_mode hybrid --prefill_ar_len 128 --max_seq_len 1024 --decoder_model qwen3-1_7b --prompt "I would like to learn python, could you teach me with a simple example?" --pre_gen_pte ./llama_qnn/

model output
Run the above script and I can get the correct output. But when I run the normal pte file and tokenizer file with the llama android demo, the result is wrong.

• android demo output:

• python shell output:

python_shell.txt

Versions

Collecting environment information...
PyTorch version: N/A
Is debug build: N/A
CUDA used to build PyTorch: N/A
ROCM used to build PyTorch: N/A

OS: Debian GNU/Linux 12 (bookworm) (x86_64)
GCC version: (Debian 12.2.0-14+deb12u1) 12.2.0
Clang version: Could not collect
CMake version: version 3.31.10
Libc version: glibc-2.3

Python version: 3.10.8 (Wed Jun 14 15:29:07 CST 2023) [Graal, GraalVM CE, Java 17.0.7] (32-bit runtime)
Python platform: Linux-5.4.143.bsk.8-amd64-x86_64-with-glibc2.3
Is CUDA available: N/A
CUDA runtime version: 12.6.85
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration:
Nvidia driver version: Could not collect
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.9.7.1
/usr/lib/x86_64-linux-gnu/libcudnn_adv.so.9.7.1
/usr/lib/x86_64-linux-gnu/libcudnn_cnn.so.9.7.1
/usr/lib/x86_64-linux-gnu/libcudnn_engines_precompiled.so.9.7.1
/usr/lib/x86_64-linux-gnu/libcudnn_engines_runtime_compiled.so.9.7.1
/usr/lib/x86_64-linux-gnu/libcudnn_graph.so.9.7.1
/usr/lib/x86_64-linux-gnu/libcudnn_heuristic.so.9.7.1
/usr/lib/x86_64-linux-gnu/libcudnn_ops.so.9.7.1
Is XPU available: N/A
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: N/A
Caching allocator config: N/A

CPU:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 46 bits physical, 57 bits virtual
Byte Order: Little Endian
CPU(s): 128
On-line CPU(s) list: 0-127
Vendor ID: GenuineIntel
Model name: Intel(R) Xeon(R) Platinum 8336C CPU @ 2.30GHz
CPU family: 6
Model: 106
Thread(s) per core: 2
Core(s) per socket: 32
Socket(s): 2
Stepping: 6
CPU(s) scaling MHz: 86%
CPU max MHz: 3500.0000
CPU min MHz: 800.0000
BogoMIPS: 4600.00
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 invpcid_single ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local wbnoinvd dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq rdpid md_clear pconfig flush_l1d arch_capabilities
Virtualization: VT-x
L1d cache: 3 MiB (64 instances)
L1i cache: 2 MiB (64 instances)
L2 cache: 80 MiB (64 instances)
L3 cache: 108 MiB (2 instances)
NUMA node(s): 4
NUMA node0 CPU(s): 0-15,64-79
NUMA node1 CPU(s): 16-31,80-95
NUMA node2 CPU(s): 32-47,96-111
NUMA node3 CPU(s): 48-63,112-127
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Enhanced IBRS, IBPB conditional, RSB filling
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected

Versions of relevant libraries:
[pip3] executorch==1.1.0a0+17adba1
[pip3] numpy==2.2.6
[pip3] nvidia-cublas-cu12==12.8.4.1
[pip3] nvidia-cuda-cupti-cu12==12.8.90
[pip3] nvidia-cuda-nvrtc-cu12==12.8.93
[pip3] nvidia-cuda-runtime-cu12==12.8.90
[pip3] nvidia-cudnn-cu12==9.10.2.21
[pip3] nvidia-cufft-cu12==11.3.3.83
[pip3] nvidia-curand-cu12==10.3.9.90
[pip3] nvidia-cusolver-cu12==11.7.3.90
[pip3] nvidia-cusparse-cu12==12.5.8.93
[pip3] nvidia-cusparselt-cu12==0.7.1
[pip3] nvidia-nccl-cu12==2.27.5
[pip3] nvidia-nvjitlink-cu12==12.8.93
[pip3] nvidia-nvtx-cu12==12.8.90
[pip3] pytorch_tokenizers==1.0.1
[pip3] torch==2.10.0
[pip3] torchao==0.15.0+git9338966da
[pip3] torchaudio==2.10.0
[pip3] torchdata==0.11.0
[pip3] torchsr==1.0.4
[pip3] torchtune==0.6.1
[pip3] torchvision==0.25.0
[pip3] triton==3.6.0
[conda] executorch 1.1.0a0+17adba1 pypi_0 pypi
[conda] numpy 2.2.6 pypi_0 pypi
[conda] nvidia-cublas-cu12 12.8.4.1 pypi_0 pypi
[conda] nvidia-cuda-cupti-cu12 12.8.90 pypi_0 pypi
[conda] nvidia-cuda-nvrtc-cu12 12.8.93 pypi_0 pypi
[conda] nvidia-cuda-runtime-cu12 12.8.90 pypi_0 pypi
[conda] nvidia-cudnn-cu12 9.10.2.21 pypi_0 pypi
[conda] nvidia-cufft-cu12 11.3.3.83 pypi_0 pypi
[conda] nvidia-curand-cu12 10.3.9.90 pypi_0 pypi
[conda] nvidia-cusolver-cu12 11.7.3.90 pypi_0 pypi
[conda] nvidia-cusparse-cu12 12.5.8.93 pypi_0 pypi
[conda] nvidia-cusparselt-cu12 0.7.1 pypi_0 pypi
[conda] nvidia-nccl-cu12 2.27.5 pypi_0 pypi
[conda] nvidia-nvjitlink-cu12 12.8.93 pypi_0 pypi
[conda] nvidia-nvtx-cu12 12.8.90 pypi_0 pypi
[conda] pytorch-tokenizers 1.0.1 pypi_0 pypi
[conda] torch 2.10.0 pypi_0 pypi
[conda] torchao 0.15.0+git9338966da pypi_0 pypi
[conda] torchaudio 2.10.0 pypi_0 pypi
[conda] torchdata 0.11.0 pypi_0 pypi
[conda] torchsr 1.0.4 pypi_0 pypi
[conda] torchtune 0.6.1 pypi_0 pypi
[conda] torchvision 0.25.0 pypi_0 pypi
[conda] triton 3.6.0 pypi_0 pypi

qnn sdk: 2.42.0
android ndk: android-ndk-r26b

cc @cccclai @cbilgin @abhinaykukkadapu

[jincan2010]

And the change of the llama demo is as below:

[metascroy]

@abhinaykukkadapu can you have a look?

[abhinaykukkadapu]

@psiddh any idea why android app gives gibberish?

@jincan2010 regarding

But when I run the normal pte file

Can you clarify what is normal pte file, and if you used the same file that is generated for app too?

[jincan2010]

@abhinaykukkadapu I used the python script below to generate the pte file which is located in the llama_qnn directory. And I used the same file for app too.

model conversion
python examples/qualcomm/oss_scripts/llama/llama.py -b build-android -m SM8550 --temperature 0 --model_mode hybrid --prefill_ar_len 128 --max_seq_len 1024 --decoder_model qwen3-1_7b --prompt "I would like to learn python, could you teach me with a simple example?" --tasks wikitext --limit 1