Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
78.4Average Performance Score
20
Apr 10, 2026
84.3Average Performance Score
20
Apr 10, 2026
85.7Average Performance Score
20
Apr 10, 2026
32.9Accuracy
20
May 21, 2026
81.6ECR
20
Mar 31, 2026
38.72F1 Score
20
Jul 3, 2026
83ASQA Score
20
Mar 5, 2026
2,513# Traces
20
Feb 26, 2026
0.3922WQA In-Domain
20
Feb 26, 2026
49.6Accuracy
20
Feb 26, 2026
50.3QA Accuracy
20
Feb 26, 2026
58.4EM
20
Feb 26, 2026
92.14Composite Score
20
Feb 26, 2026
97.8Performance
20
Feb 26, 2026
67.23Answerable EM
20
Feb 26, 2026
28.92Answerable EM
20
Feb 26, 2026
88% True (GPT-Judge)
20
Feb 26, 2026
98.86Accuracy (amb)
20
Feb 26, 2026
89.7LLM-Acc
20
Feb 26, 2026
94.5LLM Accuracy
20
Feb 26, 2026
70.6EM
20
Feb 26, 2026
45.67Exact Match
20
Feb 26, 2026
72.74EM
20
Feb 26, 2026
232.58TPS
20
Feb 26, 2026
100Attack Success Rate (ASR)
20
Feb 26, 2026