pytorch/executorch

View on GitHub

On-device AI across mobile, embedded and edge for PyTorch

Stars: 5k
Forks: 1.2k
Open beginner issues: 0
Indexed issues: 851
Avg merge: 2d 10h
Merged PRs (30d): 581

Dominant language: Python
License: No license data
Last GitHub push: Sep 20, 2026
Latest indexed: Sep 19, 2026
Contributing guide: Contributing guide
Code of conduct: Code of conduct
Beginner labels: good first issue help wanted

Beginner friendly Everything

Open All

851 open issues indexed

vit executorch inference speed much slower than onnx Open

module: xnnpack need-user-input triaged

Difficulty 4/5 3-5 days Newbie friendliness 42/100

pytorch/executorch#6961 · 12 comments · Opened Nov 19, 2024
Auto-assign reviewers for delegate PRs Open

actionable enhancement module: dev infra triaged

Difficulty 5/5 Over a week Newbie friendliness 25/100

pytorch/executorch#6911 · 1 comment · 1 reaction · Opened Nov 15, 2024
Load Llava model error: Fatal signal 11 (SIGSEGV), code 2 (SEGV_ACCERR) Open

module: examples module: llm triaged

pytorch/executorch#6887 · 3 comments · 1 reaction · 1 assignee · Opened Nov 15, 2024
Try renaming `shim/` to make it more clear that it's an internal/infra concept Open

actionable module: build/install triaged

Difficulty 5/5 Over a week Newbie friendliness 28/100

pytorch/executorch#6862 · 1 reaction · Opened Nov 14, 2024
meta-llama/Llama-3.2-1B, The model is not working properly Open

module: examples module: qnn triaged

pytorch/executorch#6848 · 5 comments · 1 assignee · Opened Nov 14, 2024
How to Apply Different Quantization Settings Per Layer in ExecuTorch? Open

module: quantization partner: qualcomm triaged

pytorch/executorch#6846 · 9 comments · 1 assignee · Opened Nov 14, 2024
How to convert tokenizer of SmolLM model as accepted by executorch Open

module: extension module: user experience triaged

pytorch/executorch#6813 · 7 comments · 2 assignees · Opened Nov 13, 2024
ConstraintViolationError Open

module: exir triaged

Difficulty 4/5 3-5 days Newbie friendliness 25/100

pytorch/executorch#6782 · 30 comments · Opened Nov 12, 2024
Error Compiling SpinQuant in QnnBackend Open

module: qnn partner: qualcomm triaged

pytorch/executorch#6748 · 7 comments · 1 assignee · Opened Nov 11, 2024
[Quantization Size Issue] Regarding the point at which the model size changes during quantization. Open

module: qnn need-user-input triaged

Difficulty 5/5 Over a week Newbie friendliness 20/100

pytorch/executorch#6711 · 2 comments · Opened Nov 7, 2024
[QNN ISSUE] QNN 16a8w Quantization Method for the Llama-3.2 Model. Open

module: qnn triaged

Difficulty 5/5 Over a week Newbie friendliness 35/100

pytorch/executorch#6684 · 5 comments · 1 reaction · Opened Nov 6, 2024
Model lowers to CoreML but fails to run Open

module: coreml need-user-input triaged

Difficulty 4/5 3-5 days Newbie friendliness 25/100

pytorch/executorch#6663 · 3 comments · Opened Nov 5, 2024
How To Building and Running Llama 3.2 1B Instruct with Qualcomm AI Engine Direct Backend？ Open

module: llm module: qnn partner: qualcomm triaged

Difficulty 4/5 3-5 days Newbie friendliness 30/100

pytorch/executorch#6655 · 6 comments · Opened Nov 5, 2024
Question Regarding Code Execution in the Calibration Process. Open

module: quantization triaged

pytorch/executorch#6629 · 8 comments · 1 assignee · Opened Nov 3, 2024
Android: Tensor type is not very friendly to BFloat16 Open

module: android triaged

pytorch/executorch#6571 · 6 comments · 2 reactions · 1 assignee · Opened Oct 29, 2024
RFC: extension/llm API design Open

rfc triaged

Difficulty 5/5 Over a week Newbie friendliness 25/100

pytorch/executorch#6558 · 4 reactions · Opened Oct 29, 2024
qd8-f16 quantization in xnnpack Open

module: xnnpack need-user-input triaged

Difficulty 5/5 Over a week Newbie friendliness 25/100

pytorch/executorch#6510 · 2 comments · Opened Oct 26, 2024
C++ includes require that the repo lives in a directory named exactly `executorch` Open

triaged

Difficulty 5/5 Over a week Newbie friendliness 35/100

pytorch/executorch#6475 · 1 comment · Opened Oct 24, 2024
Build Android aar failed due to lacking crypt.h Open

module: qnn partner: qualcomm triaged

pytorch/executorch#6454 · 4 comments · 1 reaction · 2 assignees · Opened Oct 23, 2024
build error when generate llama2 exe file in x86 computer Open

module: llm triaged

pytorch/executorch#6431 · 5 comments · 2 assignees · Opened Oct 22, 2024
Error from LLaMA 3.2 3B Instruct Model generation (.pte) Open

module: qnn partner: qualcomm triaged

pytorch/executorch#6388 · 19 comments · 1 assignee · Opened Oct 21, 2024
Cannot using examples.models.llama.export_llama due to incorrect deprecated Open

module: llm triaged

pytorch/executorch#6375 · 3 comments · 1 assignee · Opened Oct 20, 2024
Failed to run qnn_inference_runner on an 8 Gen 2 device Open

module: qnn partner: qualcomm triaged

pytorch/executorch#6188 · 4 comments · 1 assignee · Opened Oct 14, 2024
Unable to compile android runtime executable file Open

module: build/install triaged

Difficulty 4/5 3-5 days Newbie friendliness 42/100

pytorch/executorch#6086 · 2 comments · Opened Oct 10, 2024
[QCOMM] Add SoC support for Samsung flip phone Open

module: qnn partner: qualcomm triaged

pytorch/executorch#6024 · 5 comments · 1 assignee · Opened Oct 8, 2024
llama3.2 1B model run on QNN backend produce wrong result Open

module: qnn partner: qualcomm triaged

pytorch/executorch#5929 · 16 comments · 1 assignee · Opened Oct 7, 2024
QNN backend model convert cost many memory. Open

module: qnn partner: qualcomm triaged

pytorch/executorch#5926 · 5 comments · 1 assignee · Opened Oct 7, 2024
QNN GPU or DSP backend issue Open

module: qnn partner: qualcomm triaged

pytorch/executorch#5914 · 8 comments · 1 assignee · Opened Oct 5, 2024
Slow model loading time for CoreML quantized model Open

module: coreml triaged

pytorch/executorch#5718 · 15 comments · 2 assignees · Opened Sep 27, 2024
Support for LLaVA-NeXT Open

module: runtime triaged

pytorch/executorch#5712 · 1 comment · 2 assignees · Opened Sep 26, 2024
Partial weight loading for reduced RAM utilization Open

module: runtime triaged

pytorch/executorch#5626 · 2 comments · 1 assignee · Opened Sep 25, 2024
phi-3-mini is not a valid Python package name Open

module: examples triaged

pytorch/executorch#5509 · 4 comments · 1 assignee · Opened Sep 20, 2024
Operator torch._ops.aten.unfold.default is not Aten Canonical Open

module: exir triaged

Difficulty 4/5 3-5 days Newbie friendliness 35/100

pytorch/executorch#5381 · 15 comments · Opened Sep 15, 2024
The calculation results of RoPE implemented by executorch and C++ are different. Open

module: examples module: llm triaged

Difficulty 4/5 3-5 days Newbie friendliness 35/100

pytorch/executorch#5170 · 2 comments · Opened Sep 9, 2024
RuntimeError: mmap can only be used with files saved with torch.save and _use_new_zipfile_serialization=True Open

module: llm/evaluation need-user-input triaged

Difficulty 3/5 1-2 days Newbie friendliness 42/100

pytorch/executorch#5030 · 10 comments · 1 reaction · Opened Sep 3, 2024
Call for Action: Supporting Dim Order in Backend Delegates Open

triaged

pytorch/executorch#4873 · 10 comments · 5 reactions · 1 assignee · Opened Aug 23, 2024
Xtensa platform does not support pread() Open

module: cadence triaged

Difficulty 4/5 3-5 days Newbie friendliness 35/100

pytorch/executorch#4812 · 2 comments · 1 reaction · Opened Aug 20, 2024
[Llama] [Dynamic Shape] [Core ML Delegate] Do Not Delegate Symbol Manipulation Open

module: exir triaged

Difficulty 4/5 3-5 days Newbie friendliness 35/100

pytorch/executorch#4658 · 19 comments · Opened Aug 10, 2024
Option to build without the standard library Open

module: build/install module: runtime triaged

Difficulty 5/5 Over a week Newbie friendliness 30/100

pytorch/executorch#4561 · 7 comments · Opened Aug 6, 2024
Error Saving YOLOv8 Model After CoreML Conversion Open

module: coreml need-user-input triaged

pytorch/executorch#4512 · 5 comments · 1 assignee · Opened Aug 1, 2024
Failures/Errors in unit test (with QualComm's AI SDK) Open

partner: qualcomm triaged

pytorch/executorch#4191 · 4 comments · 3 reactions · 1 assignee · Opened Jul 9, 2024
llama model is not fully lowered to ANE (coreml backend) Open

module: coreml triaged

Difficulty 4/5 3-5 days Newbie friendliness 35/100

pytorch/executorch#4091 · Opened Jun 29, 2024
Failed to load data for backend XnnpackBackend Open

module: examples triaged

pytorch/executorch#3848 · 6 comments · 2 assignees · Opened Jun 5, 2024
Enable dim order in Cadence export Open

triaged

pytorch/executorch#3704 · 1 assignee · Opened May 22, 2024
Issues lowering attention module to edge Open

module: exir triaged

Difficulty 4/5 3-5 days Newbie friendliness 25/100

pytorch/executorch#3672 · 1 comment · Opened May 19, 2024
llama2 '8da4w-gptq' quantization fails Open

module: quantization triaged

pytorch/executorch#3632 · 4 comments · 1 assignee · Opened May 16, 2024
kv cache manipulation? Open

enhancement module: llm triaged

pytorch/executorch#3518 · 21 comments · 3 reactions · 1 assignee · Opened May 6, 2024
checkpoint str has no attribute 'get' Open

module: coreml triaged

pytorch/executorch#3444 · 8 comments · 1 reaction · 1 assignee · Opened May 1, 2024
Debugging to_executorch() call Open

module: devtools triaged

Difficulty 4/5 3-5 days Newbie friendliness 35/100

pytorch/executorch#1890 · 8 comments · Opened Feb 8, 2024
support for quallcomm AI 100 hardware Open

enhancement partner: qualcomm triaged

Difficulty 5/5 Over a week Newbie friendliness 25/100

pytorch/executorch#1649 · 1 comment · 2 reactions · Opened Jan 19, 2024

pytorch/executorch

Get new issues in your inbox