-
Notifications
You must be signed in to change notification settings - Fork 1.1k
All issues
Issue creation is restricted in this repository
- #7679 · mergennachin opened
on Jan 15, 2025 - #20552 · rascani opened
on Jun 26, 2026 5
Issues
is:issue state:open
is:issue state:open
Search results
Vulkan: conv2d_local_wg_size() classifies by shader name only, making its SlidingWindow branch unreachable
module: vulkanIssues related to the Vulkan delegate and code under backends/vulkan/Issues related to the Vulkan delegate and code under backends/vulkan/Status: Open.#21942 In pytorch/executorch;Vulkan: multi-stage CNNs give different results on every run (EfficientNetV2-S, YOLO26) on Adreno
module: vulkanIssues related to the Vulkan delegate and code under backends/vulkan/Issues related to the Vulkan delegate and code under backends/vulkan/Status: Open.#21939 In pytorch/executorch;Vulkan: conv2d im2col/GEMM path is non-deterministic for c_out in 168..192 (Adreno)
module: vulkanIssues related to the Vulkan delegate and code under backends/vulkan/Issues related to the Vulkan delegate and code under backends/vulkan/Status: Open.#21938 In pytorch/executorch;AtenToCortexMPass crashes with "unsupported param type, call_function" when aten.linear.default is in preserve_ops
module: microcontrollersFor embedded MCUs like Cortex-M, or RTOS like Zephyr, does not track NPU backend like Arm Ethos.For embedded MCUs like Cortex-M, or RTOS like Zephyr, does not track NPU backend like Arm Ethos.partner: armFor backend delegation, kernels, demo, etc. from the 3rd-party partner, ArmFor backend delegation, kernels, demo, etc. from the 3rd-party partner, ArmStatus: Open.#21904 In pytorch/executorch;Docs for Qualcomm scripts use outdated (removed) CLI flags
module: qnnIssues related to Qualcomm's QNN delegate and code under backends/qualcomm/Issues related to Qualcomm's QNN delegate and code under backends/qualcomm/Status: Open.#21880 In pytorch/executorch;QNN DeepLabV3 example opens too many files
module: qnnIssues related to Qualcomm's QNN delegate and code under backends/qualcomm/Issues related to Qualcomm's QNN delegate and code under backends/qualcomm/Status: Open.#21870 In pytorch/executorch;TextLLMRunner degenerates on an LFM2.5 export that steps correctly token by token
module: llmIssues related to LLM examples and apps, and to the extensions/llm/ codeIssues related to LLM examples and apps, and to the extensions/llm/ codeStatus: Open.#21858 In pytorch/executorch;Core ML
coreml_quantizefails on tied embeddings: "compression config conflict detected between ops"module: coremlIssues related to Apple's Core ML delegation and code under backends/apple/coreml/Issues related to Apple's Core ML delegation and code under backends/apple/coreml/Status: Open.#21856 In pytorch/executorch;export_llmCore ML builds fail at execute: "must be an MLState" even thoughtake_over_mutable_buffer=Falsemodule: coremlIssues related to Apple's Core ML delegation and code under backends/apple/coreml/Issues related to Apple's Core ML delegation and code under backends/apple/coreml/module: llmIssues related to LLM examples and apps, and to the extensions/llm/ codeIssues related to LLM examples and apps, and to the extensions/llm/ codepartner: appleFor backend delegation, kernels, demo, etc. from the 3rd-party partner, AppleFor backend delegation, kernels, demo, etc. from the 3rd-party partner, AppleStatus: Open.#21855 In pytorch/executorch;Runtime accepts a channels-last input for a contiguous method and reads it as contiguous (silent wrong output)
module: runtimeIssues related to the core runtime and code under runtime/Issues related to the core runtime and code under runtime/triage reviewItems require an triage reviewItems require an triage reviewStatus: Open.#21837 In pytorch/executorch;XNNPACK: quantized slice fails to build when its output scale differs from its input
module: quantizationIssues related to quantizationIssues related to quantizationmodule: xnnpackIssues related to xnnpack delegation and the code under backends/xnnpack/Issues related to xnnpack delegation and the code under backends/xnnpack/Status: Open.#21829 In pytorch/executorch;[Muse Glimmer][CUDA] SM80+PTX vision kernel exceeds RTX 5090 shared-memory limit; native SM120 export OOMs
module: cudaIssues related to the AOTI CUDA backendIssues related to the AOTI CUDA backendmodule: llmIssues related to LLM examples and apps, and to the extensions/llm/ codeIssues related to LLM examples and apps, and to the extensions/llm/ codeStatus: Open.#21827 In pytorch/executorch;