Parsimony benchmark measures how much code AI models add

The open-source Parsimony benchmark measures net code units added by AI models, not just solve rate or token cost, to capture code bloat in long-term projects. It counts syntax elements in public patches from DeepSWE, SWE-bench Verified and Live, subtracting deletions; since GPT-6 and Claude 5.5 lack public patches, the author ran a few tasks himself, shown unranked.

• Community discusses Qwen27b versus GPT 62026-10-31
• Unitree Founder: GPT-6 May Bring Earlier Robotics Breakthrough2026-10-12
• Researchers Let GPT-6 and Rivals Drive a Toyota Corolla2026-10-07
• OpenAI launches Intelligent UI, a more visual ChatGPT interface2026-10-07
• SWE-Race benchmark tests coding agents on real concurrency bugs2026-10-06