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

Long-Context Language Modeling on RULER (4K to 128K Context Sweep)

89.1Accuracy

Full (exact)

0.723.6546.669.55Apr 7, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.04
89.1--------
2026.04
88.9--------
2026.04
88.6--------
2026.04
88.4--------
2026.04
88.4--------
2026.04
88.2--------
2026.04
88--------
2026.04
84.6--------
2026.04
84.2--------
2026.04
83.3--------
2026.04
80.5--------
2026.04
79.2--------
2026.04
79--------
2026.04
78.7--------
2026.04
78.3--------
2026.04
78.1--------
2026.04
77.6--------
2026.04
77.4--------
2026.04
76.1--------
2026.04
72.1--------
2026.04
69.9--------
2026.04
68.1--------
2026.04
60--------
2026.04
58.1--------
2026.04
54.1--------
2026.04
52.5--------
2026.04
51.2--------
2026.04
47.7--------
2026.04
47--------
2026.04
42.6--------
2026.04
40.8--------
2026.04
39--------
2026.04
35.8--------
2026.04
35.4--------
2026.04
32--------
2026.04
30.2--------
2026.04
29.5--------
2026.04
23.9--------
2026.04
22.6--------
2026.04
22.6--------
2026.04
20.5--------
2026.04
15.7--------
2026.04
10.8--------
2026.04
7--------
2026.04
6.4--------
2026.04
6--------
2026.04
5.8--------
2026.04
5.5--------
2026.04
4.9--------
2026.04
4.6--------
2026.04
4.1--------
2026.01
-1.081.051.041.031.011.01--
2026.01
-0.850.660.580.530.510.5--
2026.01
-2.093.38------
2026.01
-1.060.920.820.610.670.79--
2026.01
-1.110.740.570.490.450.44--
2026.01
-1.651.170.920.650.690.8--
2026.01
-1.071.041.031.021.011.01--
2026.01
-0.870.690.610.560.550.54--
2026.01
-1.943.15------
2026.01
-1.121.030.930.830.790.81--
2026.01
-1.190.830.660.570.550.53--
2026.01
-1.731.431.251.480.790.95--
2025.12
--90.9790.186.1783.0679.6585.99-
2025.12
--21.1716.115.92.581-76.64-
2025.12
--88.8186.0181.1675.6370.48-5.58-
2025.12
--85.048580.3475.2767.51-7.36-
2025.12
--90.0688.8185.7480.474.39-2.11-
2025.12
--90.189.898683.3777.34-0.65-
2025.12
--89.989.6586.4282.7174.82-1.29-
2025.12
--90.1489.9386.283.3876.6-0.74-
2025.12
--89.681.9471.566.0960.9174.01-
2025.12
--7.779.037.17.052.83-67.25-
2025.12
--84.8575.7763.2358.1752.22-7.16-
2025.12
--86.5378.6667.5662.8958.54-3.17-
2025.12
--86.9679.2568.1965.1257.69-2.57-
2025.12
--90.381.7271.6765.0258.82-0.51-
2025.12
--90.1681.4572.5764.0957.03-0.92-
2025.12
--90.1681.4572.564.5458.55-0.56-
2025.12
-31.625.6229.55.5-18.8-
2025.12
-29.928.417.69.45.9-18.2-
2025.12
-36.13329.117.79-25-
2025.12
-37.435.13321.210.4-27.4-
2025.12
-38.735.433.824.610.7-28.6-
2025.12
-42.738.633.421.710.9-29.4-
2026.03
-95.5294.0693.4388.1984.8677.1188.86100
2026.03
-95.5193.5692.791.6880.0676.6788.3655
2026.03
-94.9893.9192.6889.5182.6675.488.1927
2026.03
-95.4693.8393.6589.8281.674.3588.1226
2026.03
-95.1793.9693.7389.9282.6775.3788.4725
2026.03
-95.4493.0892.7192.1279.5773.0387.66100
2026.03
-95.0192.5691.791.68-71.8786.7376
2026.03
-95.0691.9891.1790.9377.3170.2886.1230
2026.03
-95.0392.5691.7691.7278.671.886.9135
2026.03
-95.1892.6892.1191.8578.7872.3187.1525
2026.04
-79.5178.8262.2858.3347.5723.9658.91-
2026.04
-76.8869.6944.0620.3110.314.3437.6-
2026.04
-72.817056.2519.069.387.1939.11-
2026.04
-72.2273.6164.2451.3944.122.2254.63-
2026.04
-77.1977.1963.7559.6950.6324.3858.8-