From bd91cb7fb7517d4d9a8acd11f04d40f36643801e Mon Sep 17 00:00:00 2001 From: Paul Gauthier Date: Wed, 14 Jun 2023 19:33:39 -0700 Subject: [PATCH] copy --- README.md | 39 +++++++++++++++++++-------------------- 1 file changed, 19 insertions(+), 20 deletions(-) diff --git a/README.md b/README.md index dfa3966bb..03fc64822 100644 --- a/README.md +++ b/README.md @@ -58,35 +58,34 @@ You can find more chat transcripts on the [examples page](https://aider.chat/exa ## GPT-4 vs GPT-3.5 -Aider supports -`gpt-4`, -`gpt-4-32k`, -`gpt-3.5-turbo` -and the the brand new `gpt-3.5-turbo-16k` model. -You will probably get the best results with GPT-4, -because of its large context window and -greater competance at coding -which will allow you to edit larger codebases. +Aider supports all of OpenAI's chat models, including +the the brand new `gpt-3.5-turbo-16k` model. +To use `gpt-3.5-turbo-16k`, install from GitHub per the instructions below and launch with `aider -3`. -The GPT-3.5 models are limited to editing somewhat smaller codebases. -They are less able to follow instructions for -returning code edits in a compact format, -so aider has +You will probably get the best results with one of the GPT-4 models, +because of their large context windows, +adherance to system prompt instructions and +greater competance at coding tasks. +The GPT-4 models are also able to use a +[repository map](https://aider.chat/docs/ctags.html) +to improve their ability to make changes in larger codebases. + +The GPT-3.5 models are supported more experimentally +and are limited to editing somewhat smaller codebases. +They are less able to follow instructions and +aren't able to return code edits in a compact format. +So aider has to ask GPT-3.5 to return a full copy of any code that needs to be edited. -This rapidly uses up the context window. +This rapidly uses up tokens and can hit the limits of the context window. In practice, this means you can only use `gpt-3.5-turbo` to edit files that are smaller than about 2k tokens (8k bytes). The new `gpt-3.5-turbo-16k` model should be able to edit code up to 8k tokens (32k bytes). Aider disables the [repository map feature](https://aider.chat/docs/ctags.html) -when used with either of the GPT-3.5 models. +when used with GPT-3.5 models. The `gpt-3.5-turbo` context window is too small to include a repo map. -Evaluation is needed to determine if `gpt-3.5-turbo-16k` can make use of a repo map. - -To use `gpt-3.5-turbo-16k`, install from GitHub per the instructions below and launch with `aider -3`. -For now, aider just treats `gpt-3.5-turbo-16k` like the original `gpt-3.5-turbo` with a larger -context window. +Evaluation is still needed to determine if `gpt-3.5-turbo-16k` can make use of a repo map. ## Installation