List your company

Reasoning on BBH (T, U, F Metrics)

71.5T Score (BBH)Binary
52.67657.56362.4567.337Apr 3, 2026

Evaluation results

MethodLinks
BinaryBackbone=Qwen2.5-7B-Instruct2026.0471.50.128.4
KARLBackbone=Qwen2.5-7B-Instruct2026.0467.87.424.8
PromptingBackbone=Qwen2.5-7B-Instruct2026.0467.55.826.7
PromptingBackbone=Llama-3.1-8B-Instruct2026.0464.32.133.6
BinaryBackbone=Llama-3.1-8B-Instruct2026.0463.2036.8
KARLBackbone=Llama-3.1-8B-Instruct2026.0462.511.326.2
TernaryBackbone=Qwen2.5-7B-Instruct2026.0461.515.622.9
TernaryBackbone=Llama-3.1-8B-Instruct2026.0453.428.118.5