List your company

Quantization Latency Measurement on NVFP4 Operator

0.01Latency (ms)vllm default
-0.017160.166170.34950.53283May 30, 2026

Evaluation results

MethodLinks
vllm defaultBatch Size=128, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.01—
ScaleSweepMSEBatch Size=128, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.010.97
vllm defaultBatch Size=256, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.012—
ScaleSweepBatch Size=128, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.0121.23
ScaleSweepMSEBatch Size=256, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.0131.08
vllm defaultBatch Size=512, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.016—
ScaleSweepBatch Size=256, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.0161.35
ScaleSweepMSEBatch Size=512, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.0171.03
ScaleSweepBatch Size=512, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.0211.26
vllm defaultBatch Size=1024, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.025—
ScaleSweepMSEBatch Size=1024, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.0261.03
ScaleSweepBatch Size=1024, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.0321.29
vllm defaultBatch Size=2048, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.041—
ScaleSweepMSEBatch Size=2048, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.0421.01
ScaleSweepBatch Size=2048, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.0551.33
vllm defaultBatch Size=4096, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.072—
ScaleSweepMSEBatch Size=4096, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.0720.99
ScaleSweepBatch Size=4096, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.0991.36
ScaleSweepMSEBatch Size=8192, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.1341
vllm defaultBatch Size=8192, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.135—
ScaleSweepBatch Size=8192, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.1851.38
vllm defaultBatch Size=16384, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.258—
ScaleSweepMSEBatch Size=16384, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.261.01
ScaleSweepBatch Size=16384, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.3571.38
vllm defaultBatch Size=32768, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.502—
ScaleSweepMSEBatch Size=32768, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.5021
ScaleSweepBatch Size=32768, Hidden Dimension=8192, GPU=NVIDIA RTX PRO 6000 Blackwell2026.050.6891.37