Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

LLM Inference Efficiency on Sequence prompts Medium

0.37TTFT (s)

FedRAG

-3.85624.669553.19581.7205May 25, 2026
Updated 8d ago

Evaluation Results

MethodLinks
2026.05
0.3736.51619.11139
2026.05
0.4246.99555.65139
2026.05
0.4333.05683.78139
2026.05
0.4420.66160.84139
2026.05
2.4510.28804.731,635
2026.05
3.519.231,316.922,707
2026.05
4.8311.791,107.612,171
2026.05
5.399.961,210.852,439
2026.05
5.9316.943,592.91,323
2026.05
6.0215.323,650.171,323
2026.05
6.0714.623,721.021,323
2026.05
6.2112.563,295.091,323
2026.05
9.2917.93,643.9676
2026.05
10.135.984,454.754,922
2026.05
14.4425.845,494.15820
2026.05
15.095.486,752.256,530
2026.05
15.24.846,345.427,334
2026.05
17.4428.675,891.63772
2026.05
19.9124.568,172.84868
2026.05
21.654.189,083.678,138
2026.05
73.471.6238,419.1315,843
2026.05
89.231.4645,815.7121,091
2026.05
92.521.2849,936.5623,715
2026.05
106.021.5253,448.8320,139