Search results

[Issue]:

Status: Open.

#4177 In ROCm/AMDMIGraphX;

· xiaokaiwang1984 opened on Jul 29, 2025

[Hardware requirement] I don't have amd hardware and linux system. Can I contribute?

Status: Open.

#4171 In ROCm/AMDMIGraphX;

· coderboyisongithub opened on Jul 26, 2025

[Issue]: Inference performance is relatively bad in FP32

Status: Open.

#4170 In ROCm/AMDMIGraphX;

· Artoriuz opened on Jul 25, 2025

Generate broadcast for matrix multiplication where one of the matrix is filled with ones

Perf Improve

Status: Open.

#4166 In ROCm/AMDMIGraphX;

· umangyadav opened on Jul 24, 2025

[Issue]: Slow model compilation

Under Investigation

Status: Open.

#4164 In ROCm/AMDMIGraphX;

· Artoriuz opened on Jul 23, 2025

Support LSE output in attention group in fuse_attention

enhancementNew feature or requestNew feature or request

Status: Open.

#4150 In ROCm/AMDMIGraphX;

· shivadbhavsar opened on Jul 18, 2025

sd21 perf run memory access failure for bs4 fp16

Status: Open.

#4148 In ROCm/AMDMIGraphX;

· ahsan-ca opened on Jul 18, 2025

~7% to 11% drops observed in the migx_onnxrt_inception_v3_benchmarks

Status: Open.

#4142 In ROCm/AMDMIGraphX;

· ahsan-ca opened on Jul 16, 2025

Error in running inceptionv3 benchmark from AMDMIGraphX examples

Status: Open.

#4141 In ROCm/AMDMIGraphX;

· ahsan-ca opened on Jul 16, 2025

Parallel For loop can not work with large tensors

bugSomething isn't workingSomething isn't working

Status: Open.

Bug

#4131 In ROCm/AMDMIGraphX;

· umangyadav opened on Jul 10, 2025

Test Epic

Status: Open.

Feature

#4129 In ROCm/AMDMIGraphX;

· amd-mwu10004 opened on Jul 9, 2025

[Issue]: MIGRAPHX_MLIR_USE_SPECIFIC_OPS with where operator and boolean inputs + slowdown

Status: Open.

#4117 In ROCm/AMDMIGraphX;

· TedThemistokleous opened on Jul 2, 2025