Search results

[infra] Shared Utilities with xilinx/mlir-aie

help wantedExtra attention is neededExtra attention is needed

Status: Open.

#126 In amd/IRON;

· hunhoffe opened on Jun 23, 2026

GitHub Issue Draft: stock IRON Llama full-ELF fails on Windows Ryzen AI during pyxrt.hw_context / add_config (Invalid offset 512)

Status: Open.

#119 In amd/IRON;

· CoLuthien opened on May 30, 2026

Re-enable tile-size-8192 GeLU test

Status: Open.

#113 In amd/IRON;

· andrej opened on May 11, 2026

ERT_CMD_STATE_ERROR on Strix Halo with firmware 1.1.2.65

Status: Open.

#103 In amd/IRON;

· nickbaxter18 opened on Apr 12, 2026

Add scale_i32_bf16 operator

Status: Open.

#99 In amd/IRON;

· albiol2004 opened on Apr 9, 2026

Add dequantize operator (int32 → bfloat16)

Status: Open.

#95 In amd/IRON;

· albiol2004 opened on Apr 9, 2026

INT8 GEMM support

Status: Open.

#93 In amd/IRON;

· albiol2004 opened on Apr 9, 2026

To do: Softmax masking should be optimized

Status: Open.

#84 In amd/IRON;

· andrej opened on Mar 23, 2026

GEMM partition_N > 1 produces incorrect results (affects Llama final vocab projection)

Status: Open.

#83 In amd/IRON;

· tonyjie opened on Mar 20, 2026

Intermittent CI failures for llama on later iterations

Status: Open.

#54 In amd/IRON;

· andrej opened on Dec 10, 2025

GGML Operation Requirements

Status: Open.

#53 In amd/IRON;

· ypapadop-amd opened on Dec 10, 2025

Clean up SwiGLU reference code weight generation

Status: Open.

#52 In amd/IRON;

· andrej opened on Dec 10, 2025