dearwolf/aider
1
0
Fork 0
mirror of https://github.com/Aider-AI/aider.git synced 2025-05-30 17:24:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 3

copy

Browse source
This commit is contained in:
Paul Gauthier 2024-05-03 15:48:43 -07:00
parent 4b0dfc3d05
commit 052df34300
2 changed files with 2 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

1
_data/leaderboard.csv
View file

@ -3,6 +3,7 @@ claude-3-opus-20240229,68.4,53.4,diff,aider --opus,0.30.1,f4b1797,5/2/24
gpt-4-0613,67.7,46.6,diff,aider -4,0.18.1,3aa17c4,12/16/23
gpt-4-turbo-2024-04-09,64.4,49.2,diff,aider --gpt-4-turbo,0.30.1,e610e5b,5/1/24
gemini-1.5-pro-latest,57.1,45.9,diff-fenced,aider --model gemini/gemini-1.5-pro-latest,0.32.0,5d32dd7,5/3/24
gpt-3.5-turbo-1106,56.1,45.5,whole,aider --model gpt-3.5-turbo-1106,0.30.1,7b14d77,4/30/24
claude-3-sonnet-20240229,54.9,43.6,whole,aider --sonnet,0.25.0,a5f8076,3/6/24
Deepseek Coder,54.5,47,whole,aider --model openai/deepseek-coder,0.30.1,c07f793,4/29/24
gpt-3.5-turbo-0125,49.6,39.8,whole,aider --35turbo,0.22.0,da14474,2/2/24

1 model second first format command version commit date
3 gpt-4-0613 67.7 46.6 diff aider -4 0.18.1 3aa17c4 12/16/23
4 gpt-4-turbo-2024-04-09 64.4 49.2 diff aider --gpt-4-turbo 0.30.1 e610e5b 5/1/24
5 gemini-1.5-pro-latest 57.1 45.9 diff-fenced aider --model gemini/gemini-1.5-pro-latest 0.32.0 5d32dd7 5/3/24
6 gpt-3.5-turbo-1106 56.1 45.5 whole aider --model gpt-3.5-turbo-1106 0.30.1 7b14d77 4/30/24
7 claude-3-sonnet-20240229 54.9 43.6 whole aider --sonnet 0.25.0 a5f8076 3/6/24
8 Deepseek Coder 54.5 47 whole aider --model openai/deepseek-coder 0.30.1 c07f793 4/29/24
9 gpt-3.5-turbo-0125 49.6 39.8 whole aider --35turbo 0.22.0 da14474 2/2/24

2
docs/leaderboard.md
View file

@ -4,7 +4,7 @@
Aider works best with LLMs which are good at *editing* code, not just good at writing
code.
Aider works with the LLM to make changes to the existing code in your git repo,
so the LLM needs to be able to specify how to edit that code.
so the LLM needs to be capable of reliably specifying how to edit that code.
Aider uses a
[code editing benchmark](https://aider.chat/docs/benchmarks.html#the-benchmark)