Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Long-context understanding on LongBench (test) (Including Efficiency Metrics)

71.4FewShot Performance

Qwen3-8B

13.47228.51143.5558.589Oct 10, 2023Mar 18, 2024Aug 26, 2024Feb 3, 2025Jul 14, 2025Dec 22, 2025Jun 1, 2026
Updated 19h ago

Evaluation Results

MethodLinks
2026.01
71.443.638.228.3-63.8-------
2023.10
70.640.746.227.25355.248.83,2833101.6--
2026.01
69.941.645.930.2-66.8-------
2023.10
69.839.943.227.45356.748.31,82266.12.6--
2026.06
69.45-----49.29----44.3846.54
2026.06
69.43-----48.89----44.2846.46
2026.06
69.13-----48.27----43.4146.17
2026.06
68.84-----48.3----43.4246.32
2026.06
68.6-----48.3----43.546.11
2026.06
68.5-----47.55----41.645.6
2026.06
68.46-----48.09----43.3446.23
2026.06
68.42-----47.59----41.9445.9
2026.01
68.440.540.326.8-66.4-------
2026.06
68.23-----47.41----42.5845.55
2023.10
68.237.642.926.949.953.446.53,42438.21.9--
2026.01
6836.439.328.9-62.7-------
2026.06
67.81-----48----43.445.84
2026.06
67.6-----47.29----41.7745.36
2026.06
67.58-----46.79----40.2345.41
2023.10
67.231.837.526.28.353.237.43,42139.21.7--
2026.01
67.138.737.626.2-60.3-------
2023.10
6739.738.726.537.854.24410,295-15.6---
2026.06
66.94-----46.01----40.7144.34
2026.06
66.73-----46.38----40.1545.11
2023.10
66.337.841.726.95352.446.31,96054.73.3--
2026.06
65.52-----45.13----40.2542.18
2026.01
65.525.223.823.1-55.9-------
2026.06
65.48-----45.38----40.2442.15
2026.06
65.42-----45.33----40.2242.12
2026.06
65.28-----45.23----40.1242.05
2026.06
65.25-----45.45----40.1842.05
2026.01
65.245.249.830.4-66.5-------
2026.06
65.12-----44.94----40.0241.95
2026.06
65.02-----45.07----39.9541.92
2026.06
65-----45.03----37.2844.31
2026.06
64.92-----45.1----40.0541.88
2026.06
64.82-----45.23----39.9541.85
2026.06
64.5-----44.46----37.4243.98
2026.06
64.38-----44.86----39.8541.72
2026.06
64.18-----44.59----38.6540.72
2026.06
64.15-----44.71----39.6841.42
2026.06
64.05-----44.63----39.6241.55
2026.06
63.78-----44.42----39.5541.42
2026.06
63.71-----45.5----39.2444.47
2026.06
63.66-----46.08----40.8344.5
2026.06
63.52-----44.38----39.2241.15
2026.06
63.48-----43.64----37.5239.85
2023.10
63.435.337.426.75134.541.43,39937.72--
2023.10
63.434.538.626.849.637.641.73,421313.31.2--
2026.06
63.1-----44.11----38.8540.88
2026.06
62.85-----43.24----36.939.48
2026.06
62.15-----42.9----36.4539.22
2026.06
61.85-----43.16----37.1540.02
2026.06
61.26-----44.28----37.7843.18
2023.10
61.222.432.124.510.456.834.61,95055.92.6--
2026.01
60.541.544.228.1-64.9-------
2026.01
60.329.124.659.4-42.5-------
2026.06
59.57-----43.21----35.9842.85
2026.06
58.42-----42.21----34.4441.83
2023.10
57.932.334.325.345.148.940.63,41737.52.1--
2026.06
57.2-----40.86----33.8238.15
2026.01
55.924.720.326.3-50.1-------
2026.01
54.136.839.525.2-58.4-------
2026.01
5428.831.123.8-32-------
2026.01
50.418.422.921.6-34.6-------
2023.10
40.715.631.315.61.636.223.5214481.69.8--
2023.10
32.434.336.324.726.324.829.81,991510.41.5--
2023.10
23.823.339.22527.553.1323,328350.60.3--
2023.10
23.533.835.925.91817.825.81,94754.83.4--
2026.01
22.93019.123.8-51.6-------
2023.10
19.530.129.421.212.429.123.61,98554.63.4--
2023.10
15.716.234.824.48.449.224.81,925547.10.3--