dearwolf/aider
1
0
Fork 0
mirror of https://github.com/Aider-AI/aider.git synced 2025-05-31 17:55:01 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 9

cleanup

Browse source
This commit is contained in:
Paul Gauthier 2024-09-26 11:29:39 -07:00
parent bf015e27d6
commit 7feaccd35a
2 changed files with 6 additions and 20 deletions
Download patch file Download diff file Expand all files Collapse all files

10
aider/coders/ask_prompts.py
View file

@ -4,13 +4,9 @@ from .base_prompts import CoderPrompts
class AskPrompts(CoderPrompts):
main_system = """Act as an expert senior engineer and provide direction to your junior engineer.
Study the change request and the current code.
Describe how to modify the code to complete the request.
Explain all needed code changes clearly and completely, but concisely.
The junior engineer will rely solely on your instructions, so make them unambiguous and complete.
Always reply in the same language as the change request.
main_system = """Act as an expert code analyst.
Answer questions about the supplied code.
Always reply to the user in the same language they are using.
"""
example_messages = []

16
aider/website/_posts/2024-09-26-senior-junior.md
View file

@ -155,15 +155,15 @@ They are strong at reasoning, but often fail to output well formed
code editing instructions.
It helps to instead let them describe the solution
however they prefer and then pass that output to a more traditional LLM.
The traditional LLM can then
produce the specific code edits needed to update
This Junior LLM can then interpret the solution description and
produce the code editing instructions needed to update
the existing source code file.
Traditional frontier models like gpt-4o and Sonnet also
seem to benefit from separating code reasoning and editing.
A pair of gpt-4o
or a pair of Sonnet models
in Senior/Junior configuration outperform their previous benchmark results.
in Senior/Junior configuration outperform their previous solo benchmark results.
Another reason why this approach is newly viable is that the
speed and costs of frontier models have been rapidly improving.
@ -185,16 +185,6 @@ Some noteworthy observations:
- Pairing Sonnet/Sonnet and GPT-4o/GPT-4o provides significant lift for both models, especially for GPT-4o.
- Deepseek is surprisingly effective as a Junior model. It seems remarkably capable at turning proposed coding solutions into new, updated versions of the source files. Using the efficient "diff" editing format, Deepseek helps all the Senior models except for Sonnet.
## Related work
This approach is somewhat similar to
[Cursor's "Instant Apply"](https://fireworks.ai/blog/cursor) feature.
The main differences are:
- Aider can flexibly use any off the shelf model as the Junior.
- Aider' Junior model can use the efficient "diff" editing format to specify source code changes as a series of search/replace operations. Cursor's instant apply models essentially use the "whole" edit format, asking the model to output a full, updated copy of each edited source file.
- Cursor's apply model is highly optimized for speed and reaches 1,000 tokens/second, which mitigates the delays associated with outputting whole copies of edited files.
## Try it
Aider has built in defaults to support Senior/Junior coding with