List your company

Mathematical Reasoning on GSM8K MINCE Calibration n=400

74.5AccuracyPhi-4-mini-Inst
23.0236.38549.7563.115Jun 22, 2026

Evaluation results

MethodLinks
Phi-4-mini-InstPrecision=BF162026.0674.51.16
Llama-3.1-8B-InstPrecision=BF162026.06712.54
Qwen3-8BPrecision=BF162026.0655.250.85
Qwen3.5-35B-A3BPrecision=BF162026.06540.44
Qwen2.5-3B-InstPrecision=BF162026.06441.56
Mistral-7B-InstPrecision=BF162026.0640.752.62
Llama-3.1-8B (base)Precision=BF162026.06250.63