AI Engineering
May 7, 2026 · Artificial Intelligence
Can Large Language Models Rebuild Complex Systems? ProgramBench’s Harsh Verdict
A Stanford NLP benchmark called ProgramBench tested 200 real‑world codebases and found that current large language models, including Claude and GPT‑5, achieve near‑zero success in reconstructing full systems like SQLite, FFmpeg, and a PHP compiler from binaries alone.
AI evaluationProgramBenchcode generation benchmark
0 likes · 4 min read