Question Answering on MultiExpertQA All (test)
0.79F1 ScoreLlama 2 - 70B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Llama 2 - 70BModel size=70B, Hardware=NVIDIA A40 GPU, CPU offloading=true2025.11 | 0.79 | 892.07 | |
| Comp-LLMModel size=35B, Hardware=NVIDIA A40 GPU2025.11 | 0.78 | 108.78 | |
| Llama 2 - 13BModel size=13B, Hardware=NVIDIA A40 GPU2025.11 | 0.56 | 4.64 | |
| Llama 2 - 7BModel size=7B, Hardware=NVIDIA A40 GPU2025.11 | 0.45 | 4.12 |