List your company

Functional Correctness on JSON-Bench

55.2Functional Accuracy @1LAVE
45.11247.73150.3552.969Jan 31, 2026

Evaluation results

MethodLinks
LAVEBackbone Model=DiffuC-7B, Decoding Method=LAVE2026.0155.255.556.656.7
LAVEBackbone Model=LLaDA-8B, Decoding Method=LAVE2026.015556.156.356.4
LAVEBackbone Model=Dream-7B, Decoding Method=LAVE2026.0154.454.755.155.5
LAVEBackbone Model=LLaDA-1.5, Decoding Method=LAVE2026.0154.154.554.754.7
FS-CDBackbone Model=DiffuC-7B, Decoding Method=FS-CD2026.0153.754.154.154.1
FS-CDBackbone Model=LLaDA-1.5, Decoding Method=FS-CD2026.0153.153.353.653.6
IG-CDBackbone Model=DiffuC-7B, Decoding Method=IG-CD2026.015354.755.355.3
FS-CDBackbone Model=LLaDA-8B, Decoding Method=FS-CD2026.0152.953.153.353.5
FS-CDBackbone Model=Dream-7B, Decoding Method=FS-CD2026.0152.953.153.253.2
IG-CDBackbone Model=LLaDA-8B, Decoding Method=IG-CD2026.0152.152.953.353.7
NO-CDBackbone Model=DiffuC-7B, Decoding Method=NO-CD2026.015252.953.353.7
IG-CDBackbone Model=Dream-7B, Decoding Method=IG-CD2026.0151.553.353.553.9
IG-CDBackbone Model=LLaDA-1.5, Decoding Method=IG-CD2026.0150.651.852.252.5
NO-CDBackbone Model=Dream-7B, Decoding Method=NO-CD2026.0149.850.450.750.7
NO-CDBackbone Model=LLaDA-8B, Decoding Method=NO-CD2026.0146.948.849.249.6
NO-CDBackbone Model=LLaDA-1.5, Decoding Method=NO-CD2026.0145.546.246.546.7