-
Notifications
You must be signed in to change notification settings - Fork 1.2k
All issues
Issue creation is restricted in this repository
- #7679 · mergennachin opened
on Jan 15, 2025 - #22188 · JacobSzwejbka opened
on Aug 26, 2026 5
Issues
is:issue state:open
is:issue state:open
Search results
- Status: Open.#23011 In pytorch/executorch;
[Vulkan] SDPA buffer IO uses the wrong stride when head_dim is not divisible by 4
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.#23001 In pytorch/executorch;- Status: Open.#22985 In pytorch/executorch;
Metal backend: generated kernels index a packed copy of a non-packed view out of bounds (wrong results, crash in MPSGraph)
module: metalIssues related to the AOTI Metal backendIssues related to the AOTI Metal backendtriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#22983 In pytorch/executorch;RFC: run an ExecuTorch model from Python without PyTorch
module: extensionIssues related to code under extension/Issues related to code under extension/rfcRequest for comment and feedback on a post, proposal, etc.Request for comment and feedback on a post, proposal, etc.Status: Open.#22971 In pytorch/executorch;[Vulkan] LLM decode ~12% slower than XNNPACK 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/triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#22968 In pytorch/executorch;Metal backend: ops read a view with a storage offset before the GPU has written it (silently wrong results)
module: metalIssues related to the AOTI Metal backendIssues related to the AOTI Metal backendtriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#22956 In pytorch/executorch;Metal backend: exported
.ptedoes not load outside a process that already has PyTorch's libomp (embedded AOTI object links/opt/llvm-openmp/lib/libomp.dylib)module: metalIssues related to the AOTI Metal backendIssues related to the AOTI Metal backendtriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#22955 In pytorch/executorch;Metal backend: conv2d returns wrong results (
aoti_torch_mps_convolutionignores tensor strides)module: metalIssues related to the AOTI Metal backendIssues related to the AOTI Metal backendtriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#22951 In pytorch/executorch;- Status: Open.#22845 In pytorch/executorch;
[ET-VK] Static activation quantization passes the int8 tensor to linear_q8ta_q8csw, which expects the float
module: vulkanIssues related to the Vulkan delegate and code under backends/vulkan/Issues related to the Vulkan delegate and code under backends/vulkan/triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#22775 In pytorch/executorch;[ET-VK] Conv2d with padding="valid" asserts in make_ivec2 (breaks every SigLIP/CLIP patch embed)
module: vulkanIssues related to the Vulkan delegate and code under backends/vulkan/Issues related to the Vulkan delegate and code under backends/vulkan/triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#22774 In pytorch/executorch;