List your company

Token Consumption and Cost on HotpotQA Multi-hop Reasoning

1Relative CostCoT
0.71842.61924.526.4208Mar 13, 2026

Evaluation results

MethodLinks
CoTEvaluation Model=GPT-4o-mini2026.0311,259
Analogical PromptingEvaluation Model=GPT-4o-mini2026.031.03—
TDA-RCEvaluation Model=GPT-4o-mini2026.031.14—
Role / Persona PromptingEvaluation Model=GPT-4o-mini2026.031.16—
Instruction InductionEvaluation Model=GPT-4o-mini2026.031.19—
Prompt CanvasEvaluation Model=GPT-4o-mini2026.031.19—
HoTEvaluation Model=GPT-4o-mini2026.031.26—
Self-RefineEvaluation Model=GPT-4o-mini2026.032.68—
AoTEvaluation Model=GPT-4o-mini2026.033.68—
AFlowEvaluation Model=GPT-4o-mini2026.034.59—
CoT-SCEvaluation Model=GPT-4o-mini, n=52026.035.12—
ToTEvaluation Model=GPT-4o-mini2026.036.63—
GoTEvaluation Model=GPT-4o-mini2026.036.87—
FoTEvaluation Model=GPT-4o-mini, n=82026.038.04—