Web Knowledge Retrieval on HotpotQA
89.2AccuracyGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4oEvaluation Protocol=Closed-Source API, Emission (gCO2/q)=4.52†, Throughput (Tok/s)=55.22026.03 | 89.2 | |
| Claude 3.5 SonnetEvaluation Protocol=Closed-Source API, Emission (gCO2/q)=3.85†, Throughput (Tok/s)=62.12026.03 | 88.2 | |
| Gemini 2.5 ProEvaluation Protocol=Closed-Source API, Emission (gCO2/q)=3.90†, Throughput (Tok/s)=58.42026.03 | 87.9 | |
| EcoThinkEvaluation Protocol=Adaptive Inference, Emission (gCO2/q)=1.32, Throughput (Tok/s)=148.62026.03 | 87.6 | |
| FrugalGPT (Cascade)Evaluation Protocol=Standard CoT, Emission (gCO2/q)=1.95, Throughput (Tok/s)=88.52026.03 | 84.5 | |
| Qwen-3-8B-InstructEvaluation Protocol=Standard CoT, Emission (gCO2/q)=2.12, Throughput (Tok/s)=98.52026.03 | 81.2 | |
| Llama-3.1-8B-InstructEvaluation Protocol=Standard CoT, Emission (gCO2/q)=2.15, Throughput (Tok/s)=95.82026.03 | 78.5 |