executorch good first issues (Python)

[Arm] Size-test: Run the binary on CI 4 days ago

good first issue actionable module: arm

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Enable native Kev inference with the Vulkan backend 13 days ago

good first issue module: vulkan module: examples

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

[MLX] Probe the usable Metal compiler in Apple CMake presets 21 days ago

good first issue module: build/install module: mlx bug

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

LFM2/LFM2.5: short-conv state is not cleared when a new sequence starts, so output after reset() depends on the previous prompt 24 days ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

[Core ML] Avoid unnecessary portable-kernel linkage from coremldelegate 25 days ago

good first issue module: build/install module: coreml bug

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Consolidate executor_runners 29 days ago

good first issue module: build/install

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Enable native Kev inference with the AOTI CUDA backend 29 days ago

good first issue module: examples module: cuda

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Rename `executorch` cmake target to `prim_ops_lib` 30 days ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Consolidation of Selective Build APIs for OSS about 1 month ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

[Request impl] Devtool end-to-end tests about 1 month ago

good first issue module: devtools

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Enable native Kev inference with the WebGPU backend about 1 month ago

good first issue module: examples module: webgpu

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

[Android] Use generic JNI instead of fbjni about 1 month ago

good first issue triaged module: android

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: Add MLX Op Handler for aten.flip about 1 month ago

good first issue triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

aten.bmm, aten.mul, aten._softmax doesn't get quantized/replaced with Cortex-M kernels in Transformer model about 1 month ago

good first issue triaged module: quantization module: arm bug

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Fail to allocate temp memory by exporting torch.topk about 2 months ago

good first issue module: runtime triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Python apis for loading and saving .ptd from dictionaries about 2 months ago

good first issue triaged module: training

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Optimized linear op doesn't support bias about 2 months ago

enhancement good first issue actionable module: kernels triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: Add MLX Op Handler for `aten.trunc` 2 months ago

good first issue triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Implement load_into for the shared_ptr data loader 2 months ago

good first issue module: runtime module: training

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Re-inplace slice_copy with slice 3 months ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Add dynamism tests for xnnpack model tests 3 months ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Need a feature to get etdump while running LLAMA model on qnn with qnn_llama_runner 4 months ago

good first issue module: qnn

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

XNNPACK Buck build appears not to enable KleidiAI/SME2 while CMake enables KleidiAI by default 4 months ago

help wanted module: xnnpack partner: arm

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

insert_write_back_for_buffers_pass should inject copy_ nodes at the earliest possible spot. 4 months ago

good first issue module: exir triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: Runtime MoE expert-sort for decode (MLX backend, Qwen 3.5 MoE) 4 months ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

[MLX][Good first issue] Add Q5K quantization support 4 months ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: top-k filtering for `mlx::sample` (MLX backend) 4 months ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: isolate the MLX submodule build (MLX backend) 4 months ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

[MLX] Good First Issue: Skip MoE expert sorting during decode (MLX backend, Qwen 3.5 MoE) 4 months ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

# Good First Issue: In-Model Sampling Head for the MLX Backend (Gumbel Sampling) 4 months ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

[Good first issue] Add Q4K support to MLX backend 5 months ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Add Android model E2E test 5 months ago

good first issue module: android

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Outdated versions in links on ET Doc page 6 months ago

module: doc good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Content changes required in Doc page 6 months ago

module: doc good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: Enable Gemma 4 on MLX Backend 6 months ago

good first issue triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: Add MLX Op Handler for `aten.roll` 6 months ago

good first issue triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: Add Full Integer Support for `aten.bitwise_not` 6 months ago

good first issue triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: Add Full Integer Support for `aten.bitwise_and` 6 months ago

good first issue triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: Add MLX Op Handler for `aten.bitwise_xor` 6 months ago

good first issue triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: Add MLX Op Handler for `aten.bitwise_or` 6 months ago

good first issue triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: Add MLX Op Handler for `aten.isinf` 6 months ago

good first issue triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: Add MLX Op Handler for `aten.isnan` 6 months ago

good first issue triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Good First Issue: Add MLX Op Handler for `aten.hardtanh` 6 months ago

good first issue triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Clean up pytest warnings 8 months ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Support dim order in Java Tensor 8 months ago

good first issue module: android

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

[Android] instrumentation test use models from storage, not bundled with apk 8 months ago

good first issue triaged module: android

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

[Android] Rename JNI cpp file 8 months ago

good first issue triaged module: android

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Remove the outdated warning in building from source documentation 9 months ago

module: doc good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Add test to ensure scalar_type.fbs remains in sync across the executorch repo 9 months ago

good first issue module: exir

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Add warning when users use to_edge instead of to_edge_transform_and_lower with CoreML 9 months ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Support Standalone Batch Norm 11 months ago

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Add timestamps for pte generation in CI 12 months ago

good first issue module: ci triaged

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Zeros/Ones Tensor Constructor (Android) 12 months ago

good first issue module: android

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Add dim order variant clone operator about 1 year ago

good first issue actionable module: exir

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

[Android] Add method metadata report API in Module.java about 1 year ago

good first issue triaged module: android Android_GA

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

[Android] Add a Runtime.java about 1 year ago

good first issue triaged module: android Android_GA

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

Add pybindings for text LLM runner over 1 year ago

AI Summary: The task involves creating Python bindings for a C++ API within the ExecuTorch project, specifically for its text LLM runner. This requires writing a pybind11 C++ file, integrating it into the existing CMake build system, adding the module to setup.py, and ensuring compatibility with existing .pte files and tokenizer artifacts. The goal is to allow users to run LLM models within a Python environment, such as a Jupyter notebook.

Complexity: 4/5

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

sym_max support in ExecuTorch over 1 year ago

AI Summary: Implement support for the `sym_max` operation within the ExecuTorch framework. This is necessary to enable successful graph lowering and model execution for models utilizing dynamic cache sizes, as highlighted by an issue encountered in the `optimum-executorch` project when exporting models from the Hugging Face Transformers library.

Complexity: 4/5

good first issue

pytorch/executorch

4.8K

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

Python

#deep-learning#embedded#gpu#machine-learning#mobile#neural-network#tensor

pytorch/executorch

Open Issues Need Help