Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Multi-Document Question Answering on LongBench
Loading...
43.09
HotpotQA Acc
FullKV
13.7204
21.3452
28.97
36.5948
Feb 9, 2026
HotpotQA Acc
2WikiMQA Acc
Musique Acc
Updated 4d ago
Evaluation Results
Method
Method
Links
HotpotQA Acc
2WikiMQA Acc
Musique Acc
FullKV
Model=LLaMA-3-8B-Instr...
2026.02
43.09
36.15
21.77
CompilerKV
Model=LLaMA-3-8B-Instr...
2026.02
41.37
28.48
21.25
StreamingLLM
Model=LLaMA-3-8B-Instr...
2026.02
37.84
29.02
16.77
CompilerKV
Model=Mistral-7B-Instr...
2026.02
36.42
22.91
16.47
CompilerKV
Model=InternLM-2.5-Cha...
2026.02
32.17
19.75
20.95
CompilerKV
Model=Qwen2-7B-Instruc...
2026.02
14.85
16.45
8.4
Feedback
Search any
task
Search any
task