List your company

Reasoning Question Answering on GPQA (Latency/Efficiency)

1,158Latency (NoKV, ms)NoKV
1,139.161,266.331,393.51,520.67Jun 7, 2026

Evaluation results

MethodLinks
NoKVModel=Falcon3-7B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.061,158—————
NoKVModel=Qwen2.5-7B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.061,159—————
NoKVModel=Mistral-7B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.061,257—————
NoKVModel=Llama-3-8B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.061,280—————
NoKVModel=Falcon3-10B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.061,629—————
RKSCModel=Qwen2.5-7B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.06——598—1.943
RKSCModel=Mistral-7B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.06——590—2.1323
RKSCModel=Falcon3-7B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.06——360—3.2154
RKSCModel=Falcon3-10B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.06——450—3.6372
RKSCModel=Llama-3-8B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.06——509—2.5157
vLLMModel=Qwen2.5-7B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.06—653—1.78——
vLLMModel=Mistral-7B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.06—690—1.82——
vLLMModel=Falcon3-7B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.06—637—1.82——
vLLMModel=Falcon3-10B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.06—895—1.82——
vLLMModel=Llama-3-8B Instruct, Batch size=8, Prefix length=~1,024 tok, Branching factor (t)=8, N=502026.06—708—1.81——