List your company

Knowledge Integration Quality on NQ, TruthfulQA, WoW, HotpotQA, ELI5 Aggregate

76.7Average PerformanceGuarantRAG
53.92459.83765.7571.663Apr 9, 2026

Evaluation results

MethodLinks
GuarantRAGBase Model=Qwen3-14B, Thinking Mode=true2026.0476.718.489.34.74
o12026.0476.412.8—4.85
GuarantRAGBase Model=Qwen3-14B, Thinking Mode=false2026.0475.813.290.94.72
GPT-4o2026.047514.1—4.81
GuarantRAGBase Model=Qwen3-8B, Thinking Mode=false2026.047415.788.44.68
P-RAGBase Model=Qwen3-14B, Thinking Mode=true2026.0473.825.686.94.49
P-RAGBase Model=Qwen3-14B, Thinking Mode=false2026.0472.520.488.74.47
SOLARBase Model=Qwen3-14B, Thinking Mode=true2026.0471.927.185.14.43
DRAGINBase Model=Qwen3-14B, Thinking Mode=true2026.0471.528.284.74.44
FLAREBase Model=Qwen3-14B, Thinking Mode=true2026.0471.128.784.34.41
SOLARBase Model=Qwen3-14B, Thinking Mode=false2026.0470.522.386.84.41
RQ-RAGBase Model=Qwen3-14B, Thinking Mode=true2026.0470.329.383.44.37
DRAGINBase Model=Qwen3-14B, Thinking Mode=false2026.04702386.54.42
FLAREBase Model=Qwen3-14B, Thinking Mode=false2026.0469.623.586.14.39
P-RAGBase Model=Qwen3-8B, Thinking Mode=false2026.046922.986.14.43
RQ-RAGBase Model=Qwen3-14B, Thinking Mode=false2026.0468.924.185.24.35
SelfRAGBase Model=Qwen3-14B, Thinking Mode=true2026.0468.631.281.94.34
SOLARBase Model=Qwen3-8B, Thinking Mode=false2026.0467.524.784.54.37
DA-RAGBase Model=Qwen3-14B, Thinking Mode=true2026.0467.532.580.74.28
SelfRAGBase Model=Qwen3-14B, Thinking Mode=false2026.0467.225.983.74.32
DRAGINBase Model=Qwen3-8B, Thinking Mode=false2026.0467.125.384.24.38
FLAREBase Model=Qwen3-8B, Thinking Mode=false2026.0466.725.883.74.35
DA-RAGBase Model=Qwen3-14B, Thinking Mode=false2026.0466.227.282.44.26
RQ-RAGBase Model=Qwen3-8B, Thinking Mode=false2026.0465.926.482.94.31
SelfRAGBase Model=Qwen3-8B, Thinking Mode=false2026.0464.428.181.34.28
Standard RAGBase Model=Qwen3-14B, Thinking Mode=true2026.046434.777.24.21
DA-RAGBase Model=Qwen3-8B, Thinking Mode=false2026.0463.329.580.14.22
Standard RAGBase Model=Qwen3-14B, Thinking Mode=false2026.0462.830.478.94.19
Standard RAGBase Model=Qwen3-8B, Thinking Mode=false2026.046032.676.84.15
Qwen3-14B-TBase Model=Qwen3-14B, Thinking Mode=true2026.0459.532.9—4.79
Qwen3-14BBase Model=Qwen3-14B, Thinking Mode=false2026.0457.335.8—4.76
Qwen3-8BBase Model=Qwen3-8B, Thinking Mode=false2026.0454.838.2—4.72