-
Notifications
You must be signed in to change notification settings - Fork 4.1k
Issues
is:issue state:open
is:issue state:open
Issue creation is restricted in this repository
Search results
[Perf] Add a cuDNN SDPA decode tier to the ONNX standard Attention CUDA kernel
ep:CUDAissues related to the CUDA execution providerissues related to the CUDA execution providermodel:transformerissues related to a transformer model: BERT, GPT2, Hugging Face, Longformer, T5, etc.issues related to a transformer model: BERT, GPT2, Hugging Face, Longformer, T5, etc.performanceissues related to performance regressionsissues related to performance regressionsStatus: Open.#29714 In microsoft/onnxruntime;CUDA EP aborts on a same-buffer copy of a retained in-memory initializer (ORT_ENFORCE(dst_data != src_data); CUDA-graph variant 900/901)
ep:CUDAissues related to the CUDA execution providerissues related to the CUDA execution providerStatus: Open.#29713 In microsoft/onnxruntime;- Status: Open.#29710 In microsoft/onnxruntime;
[CUDA] MatMulNBits fpA_intB path crashes with "unsupported m" on prefill for small-N int4 models
ep:CUDAissues related to the CUDA execution providerissues related to the CUDA execution providerStatus: Open.#29691 In microsoft/onnxruntime;react-native: intermittent JSI teardown segfault — cached WeakObject destroyed after runtime (~Env, AsyncWorker)
api:Javascriptissues related to the Javascript APIissues related to the Javascript APIplatform:mobileissues related to ONNX Runtime mobile; typically submitted using templateissues related to ONNX Runtime mobile; typically submitted using templateplatform:webissues related to ONNX Runtime web; typically submitted using templateissues related to ONNX Runtime web; typically submitted using templateStatus: Open.#29678 In microsoft/onnxruntime;Emit a default-visible summary when an EP places only a small fraction of the graph (silent CPU fallback)
ep:MIGraphXissues related to AMD MI GraphX execution providerissues related to AMD MI GraphX execution providerStatus: Open.#29669 In microsoft/onnxruntime;[Build] failure for onnxruntime-1.26.0 and -march=broadwell
buildbuild issues; typically submitted using templatebuild issues; typically submitted using templateStatus: Open.#29664 In microsoft/onnxruntime;[Performance] CPU MultiHeadAttention: redundant per-head mask copy makes masked attention ~1.6× slower
performanceissues related to performance regressionsissues related to performance regressionsStatus: Open.#29655 In microsoft/onnxruntime;[ep:VitisAI] WinError 1114 A dynamic link library (DLL) initialization routine failed - potential missing DLL in Vitis AI EP 1.8.63.0
ep:VitisAIissues related to Vitis AI execution providerissues related to Vitis AI execution providerStatus: Open.#29652 In microsoft/onnxruntime;[Web/WebGPU EP] First Conv reducing a >3-channel graph input miscomputes (error scales with channel count; WASM/CPU correct)
ep:WebGPUort-web webgpu providerort-web webgpu providerep:WebNNWebNN execution providerWebNN execution providerplatform:webissues related to ONNX Runtime web; typically submitted using templateissues related to ONNX Runtime web; typically submitted using templateStatus: Open.#29651 In microsoft/onnxruntime;[CUDA] MaxPool returns wrong values and indices for dilation>1 with non-zero begin padding
ep:CUDAissues related to the CUDA execution providerissues related to the CUDA execution providerStatus: Open.#29637 In microsoft/onnxruntime;- Status: Open.#29635 In microsoft/onnxruntime;