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 ...
5.7 update: Updated the code to 5.7. I will have to update the readme over time 5.6 update: Updated the code to 5.6 with many small improvements and fixes. Apologies for the lack of response to PRs ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback