List your company

LLM Inference Efficiency on HumanEval

5,233.11TTFT (ms)ReMoE
5,155.00325,682.22416,209.4456,736.6659May 26, 2026

Evaluation results

MethodLinks
ReMoEHardware=Jetson Orin NX 16GB, Backend=llama.cpp, Number of parallel requests (-np)=1, Sequence length (-n)=128, Memory mapping (--mmap)=true2026.055,233.11337.611.99
BaselineHardware=Jetson Orin NX 16GB, Backend=llama.cpp, Number of parallel requests (-np)=1, Sequence length (-n)=128, Memory mapping (--mmap)=true2026.057,185.78672.68—