List your company

Question Answering on Bamboogle 125 QA pairs

45.9EMBase Agent
16.98824.4943239.506Oct 19, 2025

Evaluation results

MethodLinks
Base AgentBackbone=Qwen-2.5-14B-Instruct2025.1045.9
Ft. AgentBackbone=Qwen-2.5-7B-Instruct2025.1038.7
Base LLMBackbone=Qwen-2.5-14B-Instruct2025.1037.9
Base LLMBackbone=Qwen-2.5-7B-Instruct2025.1032.8
Naive RAGBackbone=Qwen-2.5-7B-Instruct2025.1030.9
Base AgentBackbone=Qwen-2.5-7B-Instruct2025.1030.7
Base AgentBackbone=Mistral-NeMo-8B-Instruct2025.1029.3
Naive RAGBackbone=Qwen-2.5-14B-Instruct2025.1029.1
Ft. AgentBackbone=Qwen-2.5-3B-Instruct2025.1028.5
Base LLMBackbone=Qwen-2.5-3B-Instruct2025.1026.1
Base LLMBackbone=Mistral-NeMo-8B-Instruct2025.1024.3
Naive RAGBackbone=Mistral-NeMo-8B-Instruct2025.1020.5
Base AgentBackbone=Qwen-2.5-3B-Instruct2025.1019.5
Naive RAGBackbone=Qwen-2.5-3B-Instruct2025.1018.1