Hongxia YangandClaude Opus 4.8 c2507fb293 [ROCm] [MoE] [Perf] Shared-expert fusion for bias-routed MoE; enable on MiniMax-M3 mxfp8 model (#46545) ...

Signed-off-by: Hongxia Yang <hongxia.yang@amd.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>

2026-06-26 07:05:20 -07:00

..

deepseek_v4

[Optimization] Skip DP padding tokens in MoE (#46428)

2026-06-23 14:54:46 -07:00

minimax_m3

[ROCm] [MoE] [Perf] Shared-expert fusion for bias-routed MoE; enable on MiniMax-M3 mxfp8 model (#46545)

2026-06-26 07:05:20 -07:00

__init__.py

[Model Refactoring] Migrate DeepSeek V4 to vllm/models/ [1/N] (#43004)

2026-05-18 19:50:02 -07:00