List your company

Hallucination Detection on HaluBench (test)

86.96HELynx-8B
65.463271.044176.62582.2059Dec 31, 2025

Evaluation results

MethodLinks
Lynx-8BSize=8B2025.1286.9660.2574.364.7997.788.3768.3977.25
HDM-4B-RLSize=4B, Reasoning Mode=true2025.1283.9668.5284.5474.3992.577.7975.8879.65
o4-miniSize=-, Reasoning Mode=true2025.1283.2371.8988.976.2894.1985.9577.8882.62
GPT-4.1Size=-, Reasoning Mode=true2025.1282.9562.5860.7966.9190.7588.1775.4175.37
HDM-4B-RLSize=4B, Reasoning Mode=false2025.1282.8463.7457.8350.8584.2180.6770.7570.13
Qwen3-32BSize=32B, Reasoning Mode=false2025.1280.9257.8369.5970.6183.9483.4372.6174.13
Bespoke-MiniCheck-7BSize=7B2025.1279.7353.0850.3864.7888.164.7578.5668.48
Qwen3-32BSize=32B, Reasoning Mode=true2025.1277.6369.3686.5880.192.9987.178.1681.7
Qwen3-4BSize=4B, Reasoning Mode=true2025.1275.1561.2882.9876.1987.6177.7276.2676.74
Alignscore-LargeSize=355M2025.1274.156.7741.9350.964.1161.0565.659.21
Qwen3-4BSize=4B, Reasoning Mode=false2025.1273.6458.2556.7252.2978.2668.8366.4364.92
MiniCheck-Flan-T5-LargeSize=783M2025.1266.8553.6542.4649.7475.4871.7276.262.3
HHEMSize=110M2025.1266.5973.7240.8546.5557.4857.226157.63
FactCG-DeBERTa-v3-LargeSize=435M2025.1266.2952.2842.0348.6778.4859.566158.33