September 26, 2024

Separating code reasoning and editing

Aider now has experimental support for using two models to complete each coding task:

  • An Architect model is asked to describe how to solve the coding problem.
  • An Editor model is given the Architect’s solution and asked to produce specific code editing instructions to apply those changes to existing source files.

Splitting up “code reasoning” and “code editing” in this manner has produced SOTA results on aider’s code editing benchmark. Using o1-preview as the Architect with either DeepSeek or o1-mini as the Editor produced the SOTA score of 85%. Using the Architect/Editor approach also significantly improved the benchmark scores of many models, compared to their previous “solo” baseline scores (striped bars).

Motivation

This approach was motivated by the release of OpenAI’s o1 models. They are strong at reasoning, but often fail to output properly formatted 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. This second Editor LLM can then interpret the solution description and produce the code editing instructions needed to update the existing source code.

This approach has recently become attractive for aider due to rapid improvements in the speed and costs of frontier models. In particular, chaining older LLMs would have been quite slow and incompatible with aider’s goal of providing an interactive, pair programming AI coding experience.

Code reasoning and code editing

Normally aider asks the model to solve a coding problem in one prompt, asking the LLM to explain the solution and return a well formatted series of file edits. All of aider’s editing formats require the LLM to return source code edits in a specific text format, so that aider can process the edits and apply them to the local source files.

Because this all happens in a single prompt/response round trip to the LLM, the model has to split its attention between solving the coding problem and conforming to the edit format.

The Architect/Editor approach splits this into two inference steps, possibly using two different LLMs:

  1. Solve the coding problem (Architect).
  2. Turn the proposed solution into a series of well formed code edits (Editor).

The Architect/Editor approach allows the Architect to focus on solving the coding problem and describe the solution however comes naturally to it. Similarly, the Editor can focus all of its attention on properly formatting the edits without needing to reason much about how to solve the coding problem.

We can assign the Architect and Editor roles to LLMs which are well suited to their needs. Strong reasoning model like o1-preview make excellent Architects, while the Editor role can be assigned to an appropriate model based on cost, speed and code editing skill.

Results

The graph above and the table below show the aider’s code editing benchmark score for various combinations of Architect and Editor models.

Some noteworthy observations:

  • Pairing o1-preview as Architect with either Deepseek or o1-mini as Editor sets a SOTA significantly above the previous best score. This result is obtained with the “whole” editing format, requiring the Editor to output a full update copy of each edited source file. Both of these steps are therefore quite slow, so probably not practical for interactive use with aider.
  • Pairing OpenAI’s o1-preview with Anthropic’s Sonnet as the Editor produces the second best result. This is an entirely practical configuration for users able to work with both providers.
  • Pairing many models with themselves in the Architect/Editor configuration can provide significant benefits. Sonnet, GPT-4o and GPT-4o-mini all scored higher when used as an Architect/Editor pair.
  • Deepseek is surprisingly effective as an Editor 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 Architect models except for Sonnet.

Try it!

The development version of aider has built in defaults to support Architect/Editor coding with o1-preview, o1-mini, GPT-4o and Claude 3.5 Sonnet. Run aider with --architect or get started quickly like this:

pip install -U aider-chat

# Change directory into a git repo
cd /to/your/git/repo

# Work with Claude 3.5 Sonnet as the Architect and Editor
export ANTHROPIC_API_KEY=your-key-goes-here
aider --sonnet --architect

# Work with OpenAI models, using gpt-4o as the Editor
export OPENAI_API_KEY=your-key-goes-here
aider --4o --architect
aider --o1-mini --architect
aider --o1-preview --architect

More info

Aider has a number of “chat modes”, and “architect” is available as a new chat mode. The --architect switch is a shortcut for --chat-mode architect. For more details, see documentation on aider’s chat modes.

Full results

Below are the benchmark results using various models as the Architect, paired with various models as the Editor. Each section includes a “baseline” result, where the model works by itself in aider’s normal “code” editing mode (not as part of an Architect/Editor configuration). This “solo” baseline represents the performance previously available when using this model with aider.

Architect Editor Edit Format Pass Rate
o1-preview o1-mini whole 85.0%
o1-preview deepseek whole 85.0%
o1-preview claude-3-5-sonnet diff 82.7%
o1-preview deepseek diff 80.5%
o1-preview gpt-4o diff 80.5%
o1-preview Baseline diff 79.7%
claude-3.5-sonnet claude-3.5-sonnet diff 80.5%
claude-3.5-sonnet deepseek diff 78.9%
claude-3.5-sonnet deepseek whole 78.9%
claude-3.5-sonnet Baseline diff 77.4%
gpt-4o gpt-4o diff 75.2%
gpt-4o deepseek diff 74.4%
gpt-4o deepseek whole 73.7%
gpt-4o Baseline diff 71.4%
o1-mini deepseek whole 71.4%
o1-mini gpt-4o diff 70.7%
o1-mini deepseek diff 69.2%
o1-mini Baseline diff 61.1%
gpt-4o-mini gpt-4o-mini whole 60.2%
gpt-4o-mini Baseline whole 55.6%