List your company

Code Reasoning on CUE-Bench

1.2VT ScoreQwen2.5-0.5B
-0.0480.2760.60.924Jun 16, 2026

Evaluation results

MethodLinks
Qwen2.5-0.5BParameter count=0.5B2026.061.21610.515.68.19.410.1
Qwen2.5-Coder-0.5BParameter count=0.5B2026.060.916.310.416.27.7910.1
Qwen3-0.6BParameter count=0.6B2026.060.712.76.513.34.25.47.2
Qwen2.5-1.5BParameter count=1.5B2026.060.711.26.312.15.35.36.8
Qwen2.5-Coder-1.5BParameter count=1.5B2026.060.575.111.24.44.25.4
Qwen3-1.7BParameter count=1.7B2026.060.58.149.83.33.74.9
Qwen2.5-3BParameter count=3B2026.060.58.3410.73.55.25.3
Qwen2.5-Coder-3BParameter count=3B2026.060.55.72.68.33.13.53.9
CodeLlama-7BParameter count=7B2026.060.48.54.17.93.63.54.7
Qwen2.5-Coder-14BParameter count=14B2026.060.42.21.93.30.40.41.4
Llama-2-7BParameter count=7B2026.060.29.87.211.64.13.76.1
Qwen3-4BParameter count=4B2026.060.14.22.35.21.91.42.5
DeepSeek-Coder-6.7BParameter count=6.7B2026.060.15.42.372.21.93.2
Qwen2.5-Coder-7BParameter count=7B2026.060.14.32.55.91.61.62.7
Qwen2.5-7BParameter count=7B2026.0605.72.38.62.83.23.8
Qwen3-8BParameter count=8B2026.0604.60001.91.1