dearwolf/aider
1
0
Fork 0
mirror of https://github.com/Aider-AI/aider.git synced 2025-05-31 09:44:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 9

Update benchmarks.md

Browse source
This commit is contained in:
paul-gauthier 2023-06-30 17:06:24 -07:00 • committed by GitHub
parent 1abbba2831
commit 0062f69d9e
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
docs/benchmarks.md
View file

@ -40,7 +40,7 @@ This produced some interesting observations:
- The new June (`0613`) versions of `gpt-3.5-turbo` are worse at code editing than the older Feb (`0301`) version. This was unexpected.
- The GPT-4 models are much better at code editing than the GPT-3.5 models. This was expected.
These results agree with an intuition that I've been
The quantitative benchmark results agree with an intuition that I've been
developing about how to prompt GPT for complex tasks like coding.
You want to minimize the "cognitive overhead" of formatting the response, so that
GPT can focus on the task at hand.