List your company

Reasoning on Downstream Tasks (PiQA, Arc E, Arc C, HS, WG, BoolQ)

80.5PiQA AccuracyFP16
50.3458.176673.83Apr 9, 2026

Evaluation results

MethodLinks
FP16Model=L2-13B, Evaluation protocol=zero-shot, Calibration samples=1282026.0480.577.549.279.472.380.673.3
FP16Model=L2-7B, Evaluation protocol=zero-shot, Calibration samples=1282026.047974.546.3766977.770.4
FP16Model=L3-8B, Evaluation protocol=zero-shot, Calibration samples=1282026.047974.546.3766977.770.4
SpinQ+GPTQ+CAEModel=L2-13B, Evaluation protocol=zero-shot, Quantization=W2A4KV4, Calibration samples=1282026.0462.845.526.444.155.862.349.5
SpinQ+GPTQModel=L2-13B, Evaluation protocol=zero-shot, Quantization=W2A4KV4, Calibration samples=1282026.045939.524.940.352.661.446.3
SpinQ+GPTQ+CAEModel=L2-7B, Evaluation protocol=zero-shot, Quantization=W2A4KV4, Calibration samples=1282026.0457.735.523.6355356.643.8
SpinQ+GPTQ+CAEModel=L3-8B, Evaluation protocol=zero-shot, Quantization=W2A4KV4, Calibration samples=1282026.0457.339.524.435.752.257.844.4
SpinQ+GPTQModel=L2-7B, Evaluation protocol=zero-shot, Quantization=W2A4KV4, Calibration samples=1282026.0457.134.923.633.253.761.444
SpinQ+GPTQModel=L3-8B, Evaluation protocol=zero-shot, Quantization=W2A4KV4, Calibration samples=1282026.0451.525.525.633.151.757.640.8