List your company
ResearchFormerly WizWand

Insight-level Evaluation on SCOpE-QA Long-context RAG collection

4.48Insight ScoreINSIGHTGEN
1.92162.58583.253.9142Apr 21, 2026
Updated 6mo ago

Evaluation results

MethodLinks
INSIGHTGENTarget Model=Claude-3.5-Sonnet, Evaluator Model=Gemini-1.5-Flash2026.044.48
FAISS+COTTarget Model=Claude-3.5-Sonnet, Evaluator Model=Gemini-1.5-Flash2026.044.24
INSIGHTGENTarget Model=GPT-4o, Evaluator Model=Gemini-1.5-Flash2026.044.2
FAISSTarget Model=GPT-4o, Evaluator Model=Gemini-1.5-Flash2026.043.86
GPT+COTTarget Model=Claude-3.5-Sonnet, Evaluator Model=Gemini-1.5-Flash2026.043.86
Direct GPTTarget Model=GPT-4o, Evaluator Model=Gemini-1.5-Flash2026.043.76
GPT+COTTarget Model=GPT-4o, Evaluator Model=Gemini-1.5-Flash2026.043.72
FAISS+COTTarget Model=GPT-4o, Evaluator Model=Gemini-1.5-Flash2026.043.61
INSIGHTGENTarget Model=Claude-3.5-Sonnet, Evaluator Model=Claude-3-Sonnet2026.043.35
INSIGHTGENTarget Model=GPT-4o, Evaluator Model=Claude-3-Sonnet2026.043.31
FAISSTarget Model=Claude-3.5-Sonnet, Evaluator Model=Gemini-1.5-Flash2026.043.22
FAISS+COTTarget Model=Claude-3.5-Sonnet, Evaluator Model=Claude-3-Sonnet2026.042.95
Direct GPTTarget Model=Claude-3.5-Sonnet, Evaluator Model=Gemini-1.5-Flash2026.042.91
FAISSTarget Model=GPT-4o, Evaluator Model=Claude-3-Sonnet2026.042.69
Direct GPTTarget Model=GPT-4o, Evaluator Model=Claude-3-Sonnet2026.042.67
GPT+COTTarget Model=GPT-4o, Evaluator Model=Claude-3-Sonnet2026.042.62
GPT+COTTarget Model=Claude-3.5-Sonnet, Evaluator Model=Claude-3-Sonnet2026.042.62
FAISS+COTTarget Model=GPT-4o, Evaluator Model=Claude-3-Sonnet2026.042.41
FAISSTarget Model=Claude-3.5-Sonnet, Evaluator Model=Claude-3-Sonnet2026.042.31
Direct GPTTarget Model=Claude-3.5-Sonnet, Evaluator Model=Claude-3-Sonnet2026.042.02