List your company

Code Generation on BigCode

46.8AccuracyMCNIG
16.84824.62432.440.176Mar 18, 2026Mar 24, 2026Mar 31, 2026Apr 7, 2026Apr 14, 2026Apr 21, 2026Apr 28, 2026

Evaluation results

MethodLinks
MCNIGModel Size=8B2026.0346.8
Self-CertaintyModel=Qwen3-14B, Number of Samples=322026.0446.2
Lowest CentroidModel=Qwen3-14B, Number of Samples=322026.0446
Tail ConfidenceModel=Qwen3-14B, Number of Samples=322026.0445.4
Greedy DecodingModel=Olmo-3.1-Think, Number of Samples=322026.0444.9
Lowest CentroidModel=Olmo-3.1-Think, Number of Samples=322026.0444.5
Lowest CentroidModel=QWQ-32B, Number of Samples=322026.0444.5
ORMModel Size=8B2026.0344.1
Bottom WindowModel=QWQ-32B, Number of Samples=322026.0444.1
Bottom WindowModel=Qwen3-14B, Number of Samples=322026.0444
Self-CertaintyModel=QWQ-32B, Number of Samples=322026.0444
Pass@1Model=QWQ-32B, Number of Samples=322026.0443.9
Bottom WindowModel=Olmo-3.1-Think, Number of Samples=322026.0443.4
Greedy DecodingModel=QWQ-32B, Number of Samples=322026.0443.4
Self-CertaintyModel=Olmo-3.1-Think, Number of Samples=322026.0443.3
IGModel Size=8B2026.0343
Pass@1Model=Olmo-3.1-Think, Number of Samples=322026.0442.6
Tail ConfidenceModel=QWQ-32B, Number of Samples=322026.0442.3
Tail ConfidenceModel=Olmo-3.1-Think, Number of Samples=322026.0441.5
Pass@1Model=Qwen3-14B, Number of Samples=322026.0439.9
Greedy DecodingModel=Qwen3-14B, Number of Samples=322026.0439.8
Tail ConfidenceModel=GPT-OSS-120B, Number of Samples=322026.0436.8
Greedy DecodingModel=GPT-OSS-120B, Number of Samples=322026.0436.1
Bottom WindowModel=GPT-OSS-120B, Number of Samples=322026.0436.1
Lowest CentroidModel=GPT-OSS-120B, Number of Samples=322026.0435.8
Pass@1Model=GPT-OSS-120B, Number of Samples=322026.0435
OVM2026.0334.9
Self-CertaintyModel=GPT-OSS-120B, Number of Samples=322026.0434.9
QwenPRMModel Size=7B2026.0331.5
Granite PRM v22026.0331.4
MathShepherd2026.0329.5
ImplicitPRM2026.0329.3
Majority voting2026.0327.6
Greedy DecodingModel=Ministral-3-14B-Instruct, Number of Samples=322026.0426.3
Lowest CentroidModel=Ministral-3-14B-Instruct, Number of Samples=322026.0425.5
Single sampling2026.0323.3
Pass@1Model=Ministral-3-14B-Instruct, Number of Samples=322026.0422.9
Tail ConfidenceModel=Ministral-3-14B-Instruct, Number of Samples=322026.0419.1
Bottom WindowModel=Ministral-3-14B-Instruct, Number of Samples=322026.0419
Self-CertaintyModel=Ministral-3-14B-Instruct, Number of Samples=322026.0418