List your company

LLM Inference Throughput on LongBench

96.93Throughput (RPS)EB(k^*)
49.256461.633274.0186.3868May 30, 2026

Evaluation results

MethodLinks
EB(k^*)GPU=H200, Model=Gemma-3-1B-IT2026.0596.934.4
v1GPU=H200, Model=Gemma-3-1B-IT2026.0592.84—
v0GPU=H200, Model=Gemma-3-1B-IT2026.0592.74—
EB(k^*)GPU=RTX PRO 6000, Model=Gemma-3-1B-IT2026.0555.310.4
v1GPU=RTX PRO 6000, Model=Gemma-3-1B-IT2026.0555.08—
v0GPU=RTX PRO 6000, Model=Gemma-3-1B-IT2026.0551.09—