List your company

Query Routing on LongBench OOD v2

53QAAlways-LC (Baseline)
30.1236.064247.94May 11, 2026

Evaluation results

MethodLinks
Always-LC (Baseline)Answer Model=DeepSeek-R1 [T]2026.055310049
Always-LC (Baseline)Answer Model=Qwen3-235B [T]2026.055210054
Pre-Route (Q235B) [N]Answer Model=DeepSeek-R1 [T]2026.055141.856
Pre-Route (Q235B) [T]Answer Model=DeepSeek-R1 [T]2026.055131.165
Self-Route (Baseline)Answer Model=Qwen3-235B [T]2026.054952.252
Pre-Route (D-Q1.7B) [T]Answer Model=DeepSeek-R1 [T]2026.054821.966
Always-LC (Baseline)Answer Model=Qwen-Max [N]2026.054810059
Pre-Route (R1) [T]Answer Model=Qwen3-235B [N]2026.054723.572
Pre-Route (Q235B) [N]Answer Model=Qwen3-235B [N]2026.054737.462
Pre-Route (Q1.7B) [N]Answer Model=Qwen3-235B [N]2026.054736.761
Pre-Route (D-Q1.7B) [N]Answer Model=Qwen3-235B [N]2026.054727.967
Pre-Route (Q235B) [N]Answer Model=Qwen3-235B [T]2026.054741.355
Pre-Route (Q235B) [T]Answer Model=Qwen3-235B [T]2026.054728.561
Pre-Route (Q1.7B) [N]Answer Model=Qwen3-235B [T]2026.054737.155
Pre-Route (D-Q1.7B) [N]Answer Model=Qwen3-235B [T]2026.054728.761
Pre-Route (R1) [T]Answer Model=DeepSeek-R1 [T]2026.054727.661
Pre-Route (Q1.7B) [N]Answer Model=DeepSeek-R1 [T]2026.054742.353
Pre-Route (Q1.7B) [T]Answer Model=DeepSeek-R1 [T]2026.054741.357
Always-LC (Baseline)Answer Model=Qwen3-235B [N]2026.054610060
Pre-Route (Q1.7B) [T]Answer Model=Qwen3-235B [N]2026.054640.556
Pre-Route (D-Q1.7B) [T]Answer Model=Qwen3-235B [N]2026.054626.766
Pre-Route (Q1.7B) [T]Answer Model=Qwen3-235B [T]2026.05464355
Pre-Route (D-Q1.7B) [T]Answer Model=Qwen3-235B [T]2026.05462958
Self-Route (Baseline)Answer Model=DeepSeek-R1 [T]2026.054644.453
Pre-Route (D-Q1.7B) [N]Answer Model=DeepSeek-R1 [T]2026.054618.464
Self-Route (Baseline)Answer Model=Qwen3-235B [N]2026.054560.944
Pre-Route (Q235B) [T]Answer Model=Qwen3-235B [N]2026.054527.467
Pre-Route (R1) [T]Answer Model=Qwen3-235B [T]2026.054523.863
Always-RAG (Baseline)Answer Model=DeepSeek-R1 [T]2026.0545051
Self-Route (Baseline)Answer Model=Qwen-Max [N]2026.054557.446
Pre-Route (Q1.7B) [N]Answer Model=Qwen-Max [N]2026.054538.760
Pre-Route (D-Q1.7B) [T]Answer Model=Qwen-Max [N]2026.054527.766
Pre-Route (Q235B) [N]Answer Model=Qwen-Max [N]2026.05444059
Pre-Route (Q235B) [T]Answer Model=Qwen-Max [N]2026.05442965
Pre-Route (Q1.7B) [T]Answer Model=Qwen-Max [N]2026.054441.856
Pre-Route (D-Q1.7B) [N]Answer Model=Qwen-Max [N]2026.054427.366
Always-RAG (Baseline)Answer Model=Qwen3-235B [T]2026.0543046
Pre-Route (R1) [T]Answer Model=Qwen-Max [N]2026.054325.868
Pre-Route (Q235B) [N]Answer Model=Qwen3-30B [N]2026.05423960
Pre-Route (D-Q1.7B) [N]Answer Model=Qwen3-30B [N]2026.054227.668
Pre-Route (D-Q1.7B) [T]Answer Model=Qwen3-30B [N]2026.05422867
Always-RAG (Baseline)Answer Model=Qwen3-235B [N]2026.0542040
Self-Route (Baseline)Answer Model=Qwen3-8B [T]2026.054132.367
Pre-Route (R1) [T]Answer Model=Qwen3-30B [N]2026.054125.970
Pre-Route (Q235B) [T]Answer Model=Qwen3-30B [N]2026.054129.267
Always-LC (Baseline)Answer Model=Qwen3-8B [N]2026.054010040
Always-LC (Baseline)Answer Model=Qwen3-8B [T]2026.054010037
Pre-Route (Q235B) [N]Answer Model=Qwen3-8B [T]2026.05403363
Always-LC (Baseline)Answer Model=Qwen3-30B [N]2026.054010040
Self-Route (Baseline)Answer Model=Qwen3-30B [N]2026.05403861
Pre-Route (Q1.7B) [T]Answer Model=Qwen3-30B [N]2026.05404157
Always-RAG (Baseline)Answer Model=Qwen-Max [N]2026.0540041
Pre-Route (R1) [T]Answer Model=Qwen3-8B [T]2026.053919.971
Always-RAG (Baseline)Answer Model=Qwen3-30B [N]2026.0539060
Self-Route (Baseline)Answer Model=Qwen3-8B [N]2026.053831.364
Pre-Route (Q235B) [N]Answer Model=Qwen3-8B [N]2026.053831.864
Pre-Route (Q1.7B) [T]Answer Model=Qwen3-8B [N]2026.053838.961
Pre-Route (D-Q1.7B) [N]Answer Model=Qwen3-8B [N]2026.053820.770
Pre-Route (Q235B) [T]Answer Model=Qwen3-8B [T]2026.053829.663
Pre-Route (Q1.7B) [N]Answer Model=Qwen3-8B [T]2026.053838.158
Pre-Route (D-Q1.7B) [N]Answer Model=Qwen3-8B [T]2026.053821.968
Pre-Route (Q1.7B) [N]Answer Model=Qwen3-30B [N]2026.053839.957
Always-LC (Baseline)Answer Model=Qwen3-4B [T]2026.053710040
Self-Route (Baseline)Answer Model=Qwen3-4B [T]2026.053733.865
Pre-Route (R1) [T]Answer Model=Qwen3-8B [N]2026.053719.672
Pre-Route (Q235B) [T]Answer Model=Qwen3-8B [N]2026.053725.868
Always-RAG (Baseline)Answer Model=Qwen3-8B [T]2026.0537063
Pre-Route (Q1.7B) [T]Answer Model=Qwen3-8B [T]2026.05374054
Pre-Route (Q235B) [N]Answer Model=Qwen3-4B [T]2026.053638.759
Pre-Route (D-Q1.7B) [T]Answer Model=Qwen3-8B [N]2026.053618.969
Pre-Route (D-Q1.7B) [T]Answer Model=Qwen3-8B [T]2026.053621.268
Always-LC (Baseline)Answer Model=Qwen3-30B [T]2026.053610056
Pre-Route (R1) [T]Answer Model=Qwen3-4B [T]2026.053524.570
Pre-Route (Q1.7B) [N]Answer Model=Qwen3-8B [N]2026.053539.855
Always-RAG (Baseline)Answer Model=Qwen3-1.7B [T]2026.0534069
Always-RAG (Baseline)Answer Model=Qwen3-4B [T]2026.0534060
Pre-Route (Q235B) [T]Answer Model=Qwen3-4B [T]2026.053430.164
Pre-Route (Q1.7B) [N]Answer Model=Qwen3-4B [T]2026.053438.258
Pre-Route (Q1.7B) [T]Answer Model=Qwen3-4B [T]2026.053440.755
Pre-Route (D-Q1.7B) [N]Answer Model=Qwen3-4B [T]2026.053421.167
Pre-Route (D-Q1.7B) [T]Answer Model=Qwen3-4B [T]2026.053420.466
Always-RAG (Baseline)Answer Model=Qwen3-8B [N]2026.0534060
Always-RAG (Baseline)Answer Model=Qwen3-1.7B [N]2026.0533062
Self-Route (Baseline)Answer Model=Qwen3-1.7B [T]2026.053328.168
Pre-Route (R1) [T]Answer Model=Qwen3-1.7B [T]2026.053323.571
Pre-Route (D-Q1.7B) [N]Answer Model=Qwen3-1.7B [T]2026.0533880
Pre-Route (D-Q1.7B) [T]Answer Model=Qwen3-1.7B [T]2026.05337.580
Self-Route (Baseline)Answer Model=Qwen3-1.7B [N]2026.053234.963
Pre-Route (D-Q1.7B) [T]Answer Model=Qwen3-1.7B [N]2026.05326.682
Pre-Route (Q235B) [T]Answer Model=Qwen3-1.7B [T]2026.053229.465
Pre-Route (Q1.7B) [T]Answer Model=Qwen3-1.7B [T]2026.053238.758
Pre-Route (Q235B) [T]Answer Model=Qwen3-4B [N]2026.05322769
Pre-Route (Q1.7B) [T]Answer Model=Qwen3-4B [N]2026.053242.158
Pre-Route (Q235B) [N]Answer Model=Qwen3-1.7B [N]2026.053139.758
Pre-Route (Q235B) [T]Answer Model=Qwen3-1.7B [N]2026.053129.766
Pre-Route (D-Q1.7B) [N]Answer Model=Qwen3-1.7B [N]2026.05317.381
Pre-Route (Q235B) [N]Answer Model=Qwen3-1.7B [T]2026.053137.860
Pre-Route (Q1.7B) [N]Answer Model=Qwen3-1.7B [T]2026.053138.956
Always-RAG (Baseline)Answer Model=Qwen3-4B [N]2026.0531070
Pre-Route (R1) [T]Answer Model=Qwen3-4B [N]2026.053124.769