News for hackers. Stuff that matters.
Re:Lazy coding benchmark for GPT-4-0125-preview by anotherpaulg on Thursday January 25, 2024 @04:17PM (#39135437)
OpenAI just released a new version of GPT-4 Turbo. This new model is intended to reduce the “laziness” that has been widely observed with the previous gpt-4-1106-preview model:

Today, we are releasing an updated GPT-4 Turbo preview model, gpt-4-0125-preview. This model completes tasks like code generation more thoroughly than the previous preview model and is intended to reduce cases of “laziness” where the model doesn’t complete a task.

With that in mind, I am benchmarking the new model using aider’s existing lazy coding benchmark. I've shared some preliminary results.

Overall, the new gpt-4-0125-preview model does worse on the lazy coding benchmark as compared to the November gpt-4-1106-preview model:

1. It performs much worse when using the unified diffs code editing format.

2. Using aider’s older SEARCH/REPLACE block editing format, the new January model outperforms the older November model. But it still performs worse than both models using unified diffs.

Re:Lazy coding benchmark for GPT-4-0125-preview by freediver on Thursday January 25, 2024 @04:29PM (#39135593)
Thanks for your work on Aider and the useful benchmark posts.