List your company

Diffusion Acceleration Code Generation on DiffBench (test)

90.24Level 1 AccuracyClaude Sonnet 4 w/ DiffAgent
36.971250.800664.6378.4594Jan 6, 2026

Evaluation results

MethodLinks
Claude Sonnet 4 w/ DiffAgentBase Model=Claude Sonnet 4, Agent Framework=DiffAgent, Evaluation Protocol=augmented2026.0190.2491.3899.2333.3363.4481.59
Claude Sonnet 4Base Model=Claude Sonnet 4, Agent Framework=None, Evaluation Protocol=out-of-the-box2026.0178.0472.4176.255.388.654.3
o3-mini w/ DiffAgentBase Model=o3-mini, Agent Framework=DiffAgent, Evaluation Protocol=augmented2026.0173.1770.6969.7322.5827.9656.46
GPT-4.1Base Model=GPT-4.1, Agent Framework=None, Evaluation Protocol=out-of-the-box2026.0156.118.977.2810.7512.914.24
o3-miniBase Model=o3-mini, Agent Framework=None, Evaluation Protocol=out-of-the-box2026.0141.4624.144.69.686.4511.92
Gemini 2.5 FlashBase Model=Gemini 2.5 Flash, Agent Framework=None, Evaluation Protocol=out-of-the-box2026.0139.0229.317.662.151.0812.09