List your company

Code Generation on SWE-Bench (Efficiency Metrics)

2.96Speedup RatioBlockPilot
1.26481.70492.1452.5851Jun 30, 2026

Evaluation results

MethodLinks
BlockPilotModel=Qwen3-8B, Temperature=02026.062.96—
DFlashModel=Qwen3-8B, Temperature=0, Block size (n)=162026.062.72—
BlockPilotModel=Qwen3-4B, Temperature=12026.062.58—
DFlashModel=Qwen3-8B, Temperature=0, Block size (n)=322026.062.5—
BlockPilotModel=Qwen3-4B, Temperature=02026.062.45—
DFlashModel=Qwen3-8B, Temperature=0, Block size (n)=82026.062.45—
DFlashModel=Qwen3-4B, Temperature=0, Block size (n)=162026.062.33—
DFlashModel=Qwen3-4B, Temperature=1, Block size (n)=162026.062.33—
BlockPilotModel=Qwen3-8B, Temperature=12026.062.3—
DFlashModel=Qwen3-4B, Temperature=1, Block size (n)=82026.062.17—
DFlashModel=Qwen3-8B, Temperature=1, Block size (n)=162026.062.09—
DFlashModel=Qwen3-8B, Temperature=1, Block size (n)=322026.062.01—
DFlashModel=Qwen3-8B, Temperature=0, Block size (n)=42026.061.96—
DFlashModel=Qwen3-4B, Temperature=0, Block size (n)=82026.061.95—
DFlashModel=Qwen3-8B, Temperature=1, Block size (n)=82026.061.94—
DFlashModel=Qwen3-4B, Temperature=1, Block size (n)=322026.061.91—
DFlashModel=Qwen3-4B, Temperature=1, Block size (n)=42026.061.79—
DFlashModel=Qwen3-4B, Temperature=0, Block size (n)=322026.061.69—
DFlashModel=Qwen3-8B, Temperature=1, Block size (n)=42026.061.66—
DFlashModel=Qwen3-4B, Temperature=0, Block size (n)=42026.061.59—
EAGLE-3Model=Qwen3-8B, Temperature=02026.061.47—
EAGLE-3Model=Qwen3-4B, Temperature=12026.061.47—
EAGLE-3Model=Qwen3-4B, Temperature=02026.061.39—
EAGLE-3Model=Qwen3-8B, Temperature=12026.061.33—