List your company

MoE LLM Serving on Qwen3-30B-A3B (BF16, top-8 experts)

26.79Throughput (tokens/s)WiSP
8.246813.060917.87522.6891Jun 20, 2026

Evaluation results

MethodLinks
WiSPSetting=SIM, GPU budget=≈67 GiB, cap amount=1282026.0626.791.01
Naive baselineSetting=SIM, GPU budget=≈67 GiB, offload amount=152026.0626.53—
WiSPSetting=SIM, GPU budget=≈44 GiB, cap amount=642026.0625.751.95
WiSPSetting=SIM, GPU budget=≈32 GiB, cap amount=322026.0618.511.8
WiSPSetting=SIM, GPU budget=≈23 GiB, cap amount=162026.0613.791.54
Naive baselineSetting=SIM, GPU budget=≈44 GiB, offload amount=302026.0613.21—
Naive baselineSetting=SIM, GPU budget=≈32 GiB, offload amount=442026.0610.27—
Naive baselineSetting=SIM, GPU budget=≈23 GiB, offload amount=502026.068.96—