List your company

Code Generation on LiveCodeBench v5 (test)

72.1Pass@1BF16 (Teacher)
24.36436.75749.1561.543May 15, 2026May 18, 2026May 21, 2026May 25, 2026May 28, 2026May 31, 2026Jun 4, 2026

Evaluation results

MethodLinks
BF16 (Teacher)sampling=8 samples per problem2026.0672.1
CKA-QADsampling=8 samples per problem2026.0670.3
NVFP4 PTQsampling=8 samples per problem2026.0668.9
QAD (KL-only)sampling=8 samples per problem2026.0668.9
CAPSModel=GPT-OSS-20B, N=16, Temperature=0.62026.0561.6
V1-InferModel=GPT-OSS-20B, N=16, Temperature=0.62026.0558.4
PointwiseModel=GPT-OSS-20B, N=16, Temperature=0.62026.0552.8
CAPSModel=Qwen3-4B-Thinking-2507, N=16, Temperature=0.62026.0551.6
VanillaModel=GPT-OSS-20B, N=16, Temperature=0.62026.0549.8
CAPSModel=Qwen3-4B-Instruct-2507, N=16, Temperature=0.62026.0548.7
V1-InferModel=Qwen3-4B-Thinking-2507, N=16, Temperature=0.62026.0545.2
V1-InferModel=Qwen3-4B-Instruct-2507, N=16, Temperature=0.62026.0542.3
PointwiseModel=Qwen3-4B-Thinking-2507, N=16, Temperature=0.62026.0539.8
PointwiseModel=Qwen3-4B-Instruct-2507, N=16, Temperature=0.62026.0538
CAPSModel=Qwen3-14B, N=16, Temperature=0.62026.0537.6
VanillaModel=Qwen3-4B-Thinking-2507, N=16, Temperature=0.62026.0536.9
VanillaModel=Qwen3-4B-Instruct-2507, N=16, Temperature=0.62026.0535.3
V1-InferModel=Qwen3-14B, N=16, Temperature=0.62026.0534.8
PointwiseModel=Qwen3-14B, N=16, Temperature=0.62026.0529.3
VanillaModel=Qwen3-14B, N=16, Temperature=0.62026.0526.2