Search results

[Perf Regression] 7 config(s) regressed @ bdcc62e9

performance

regression

Status: Open.

#990 In ROCm/ATOM;

· github-actions opened on May 31, 2026

ATOM Development Roadmap (2026 Q3)

Status: Open.

#989 In ROCm/ATOM;

· sunway513 opened on May 31, 2026

ATOM Development Roadmap (2026 Q2)

Status: Open.

#988 In ROCm/ATOM;

· sunway513 opened on May 31, 2026

[Perf Regression] 16 config(s) regressed @ 4e9160bf

performance

regression

Status: Open.

#980 In ROCm/ATOM;

· github-actions opened on May 30, 2026

[Issue]: Running DeepSeek V3.2 on MI308x*8 with Atom-vLLM only 200-300K TPM in single-machine performance

Status: Open.

#896 In ROCm/ATOM;

· LoadingZhang opened on May 24, 2026

[Issue]: readme example fails on gfx1201 (R9700)

Status: Open.

#867 In ROCm/ATOM;

· andysalerno opened on May 21, 2026

[Kernel Gap Analysis] DeepSeek-V4 AMD Porting — Block-by-Block Kernel Inventory

documentationImprovements or additions to documentationImprovements or additions to documentation

enhancementNew feature or requestNew feature or request

workstream

Status: Open.

#807 In ROCm/ATOM;

· ChuanLi1101 opened on May 15, 2026

Fail to run DeepSeek V3.2 on MI 300x

Status: Open.

#797 In ROCm/ATOM;

· LoadingZhang opened on May 15, 2026

[ATOM-Plugin][Attention] Refactor ATOM-Plugin Attention Architecture

Status: Open.

#758 In ROCm/ATOM;

· zejunchen-zejun opened on May 12, 2026

[RFC]: Enable ATOM as vLLM-Omni's out-of-tree Plugin

Status: Open.

#716 In ROCm/ATOM;

· Phi-C opened on May 8, 2026

[Tracker]: DeepSeek-V4 Support and Optimization on ATOM

workstream

Status: Open.

#707 In ROCm/ATOM;

· jamestangg opened on May 7, 2026

[Issue]: aiter ASM paged-attention with block_size=16

Status: Open.

#702 In ROCm/ATOM;

· yhl-amd opened on May 6, 2026