List your company

LLM Inference Performance on Synthetic Poisson Process Requests

18.7TPOT Speedup (%)FlashSampling
-0.4364.5329.514.468Mar 16, 2026

Evaluation results

MethodLinks
FlashSamplingB=64, Model=Qwen3-1.7B, Hardware=single B200 GPU2026.0318.7
FlashSamplingB=16, Model=Qwen3-1.7B, Hardware=single B200 GPU2026.0317.7
FlashSamplingB=8, Model=Qwen3-1.7B, Hardware=single B200 GPU2026.0315.2
FlashSamplingB=4, Model=Qwen3-1.7B, Hardware=single B200 GPU2026.0314.5
FlashSamplingB=2, Model=Qwen3-1.7B, Hardware=single B200 GPU2026.0312.8
FlashSamplingB=1, Model=Qwen3-1.7B, Hardware=single B200 GPU2026.0310.8
FlashSamplingB=32, Model=Qwen3-1.7B, Hardware=single B200 GPU2026.039.7
FlashSamplingB=2, Model=Qwen3-8B, Hardware=single B200 GPU2026.036.9
FlashSamplingB=64, Model=Qwen3-8B, Hardware=single B200 GPU2026.035.2
FlashSamplingB=32, Model=Qwen3-8B, Hardware=single B200 GPU2026.034.4
FlashSamplingB=4, Model=Qwen3-8B, Hardware=single B200 GPU2026.033.7
FlashSamplingB=8, Model=Qwen3-8B, Hardware=single B200 GPU2026.033.6
FlashSamplingB=16, Model=Qwen3-8B, Hardware=single B200 GPU2026.033
FlashSamplingB=1, Model=Qwen3-8B, Hardware=single B200 GPU2026.032.8
FlashSamplingB=4, Model=gpt-oss-120b, Hardware=single B200 GPU2026.032.4
FlashSamplingB=8, Model=gpt-oss-120b, Hardware=single B200 GPU2026.032.1
FlashSamplingB=1, Model=Qwen3-32B, Hardware=single B200 GPU2026.031.9
FlashSamplingB=2, Model=Qwen3-32B, Hardware=single B200 GPU2026.031.8
FlashSamplingB=16, Model=gpt-oss-120b, Hardware=single B200 GPU2026.031.8
FlashSamplingB=2, Model=gpt-oss-120b, Hardware=single B200 GPU2026.031.7
FlashSamplingB=4, Model=Qwen3-32B, Hardware=single B200 GPU2026.031.6
FlashSamplingB=64, Model=gpt-oss-120b, Hardware=single B200 GPU2026.031.6
FlashSamplingB=32, Model=gpt-oss-120b, Hardware=single B200 GPU2026.031.5
FlashSamplingB=32, Model=Qwen3-32B, Hardware=single B200 GPU2026.031.3
FlashSamplingB=64, Model=Qwen3-32B, Hardware=single B200 GPU2026.031.2
FlashSamplingB=8, Model=Qwen3-32B, Hardware=single B200 GPU2026.031.1
FlashSamplingB=16, Model=Qwen3-32B, Hardware=single B200 GPU2026.031.1
FlashSamplingB=1, Model=gpt-oss-120b, Hardware=single B200 GPU2026.030.3