We evaluate DeepCode on the PaperBench benchmark (released by OpenAI), a rigorous testbed requiring AI agents to independently reproduce 20 ICML 2024 papers from scratch. The benchmark comprises 8,316 ...
If you're out of time to get to the store for more wrapping paper, here's a fun tip. All you need is AI and a printer.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback