Issues

• [Issue]: fp8 mla_decode_fwd (mla_decode_stage1_asm_fwd) returns NaN / memory-faults on 512-wide (RoPE-free) MLA rows

  ROCm/aiter#6892 · Filter by author amd-sriram opened on Oct 23, 2026
  

• Add gfx950 FlyDSL 4-wave INT8 bpreshuffle GEMM

  ROCm/aiter#6827 · Filter by author ksikiric opened on Oct 22, 2026

• Fixes from running agentic RL rollouts with SGLang on MI350X / MI355X

  ROCm/aiter#6818 · Filter by author Arist12 opened on Oct 22, 2026
  1

• Update model_shapes.json to reflect new and latest models' shapes/kernels

  Triton/Gluon
  ROCm/aiter#6781 · Filter by author Boss2002n opened on Oct 21, 2026
  

• Tuning should use the benchmarks from benchmark folder

  ROCm/aiter#6777 · Filter by author Boss2002n opened on Oct 21, 2026

• remove nesting from config files

  Triton/Gluon
  ROCm/aiter#6776 · Filter by author Boss2002n opened on Oct 21, 2026

• Automate the addition of new model shapes and kernels into model shapes json in benchmarks

  Triton/Gluon
  ROCm/aiter#6775 · Filter by author Boss2002n opened on Oct 21, 2026

• track Upstream vs Internal triton regressions

  Triton/Gluon
  ROCm/aiter#6774 · Filter by author Boss2002n opened on Oct 21, 2026

• [Bug] gfx942: non-persistent bf16 MLA decode still wraps past 4 GiB (m16x4/m32x4) and ignores causal at 64/128 heads

  ROCm/aiter#6733 · Filter by author AndreasKaratzas opened on Oct 20, 2026

• [Bug] gfx942: fp8 MLA decode mla_a8w8_qh64_qseqlen4_gqaratio16_ps does not apply kv_scale to V

  ROCm/aiter#6732 · Filter by author AndreasKaratzas opened on Oct 20, 2026

• [Bug] gfx942: persistent bf16 MLA decode (mla_a16w16_qh16_m16x4_n16x1_coex0_mask1_ps) returns wrong output in some calls, nondeterministically

  ROCm/aiter#6731 · Filter by author AndreasKaratzas opened on Oct 20, 2026

• [Feature]: dynamic_mxfp4_quant(use_sr=True) outside gfx950: Triton emulation of the stochastic E2M1 conversion

  ROCm/aiter#6686 · Filter by author Ragua1 opened on Oct 19, 2026

• [Feature]: Add arch_info.is_sw_emulated_fp4_avail() for Triton MXFP4 on gfx11 GPUs without native FP4

  ROCm/aiter#6685 · Filter by author Ragua1 opened on Oct 19, 2026

• [Issue]: bench_conv2d.py: Torch column is near zero for MIOpen 3x3 convs on Windows: do_bench's 256 MiB cache clear triggers a HIP event-timing defect

  ROCm/aiter#6667 · Filter by author Ragua1 opened on Oct 19, 2026

• [Issue]: Per-tensor quantization writes past output tails for widths not divisible by four

  ROCm/aiter#6647 · Filter by author akshatvishu opened on Oct 17, 2026

• [Feature] Mamba2 selective state update (SSU) kernel

  ROCm/aiter#6644 · Filter by author AndreasKaratzas opened on Oct 17, 2026
  1

• [Feature] Non-gated ReLU² fused MoE for BF16 and FP8 weights

  ROCm/aiter#6643 · Filter by author AndreasKaratzas opened on Oct 17, 2026

• [Triton] [GFX950] mxfp4 GEMM perf worse on preshuffled vs un-shuffled

  Triton/Gluon
  Task

  ROCm/aiter#6632 · Filter by author k50112113 opened on Oct 16, 2026

• Replace the remaining hand-set checkAllclose tolerances once alignment and the measured error are known

  ROCm/aiter#6596 · Filter by author KHassanQazi opened on Oct 15, 2026

• [Feature]: [kimik3][vllm] enable router gemm aiter kernels for fp32 output in vllm

  ROCm/aiter#6591 · Filter by author nsusanto opened on Oct 15, 2026
  

• [Feature]: [Kimik3][vllm] get router gemm fp32 output enabled in VLLM

  ROCm/aiter#6590 · Filter by author nsusanto opened on Oct 15, 2026
  

• RDNA3/RDNA4 (gfx1100, gfx1201) Triton-only test report: fused KDA decode works; two install/config issues

  ROCm/aiter#6574 · Filter by author boxwrench opened on Oct 15, 2026

• [Issue]: a8w4 fused_moe writes NaN at small token counts on gfx950 (uninitialized padded scratch)

  ROCm/aiter#6565 · Filter by author PerryZhang01 opened on Oct 15, 2026
  2

• FlyDSL FP4 paged MQA-logits prefill (flydsl_pa_mqa_logits_fp4_prefill) faults at address 0 with zero query rows (gfx950)

  ROCm/aiter#6562 · Filter by author ChangLi811 opened on Oct 15, 2026

• [Issue]: FlyDSL MoE stage-1 fused FP8 output uses a different MXFP8 scale rule than fused_dynamic_mx_quant_moe_sort

  ROCm/aiter#6548 · Filter by author josh-schell-amd opened on Oct 14, 2026