From d12246c1f2cfef28efa8098c4c8980ee0dd1523b Mon Sep 17 00:00:00 2001 From: Paul Gauthier Date: Fri, 24 May 2024 07:25:40 -0700 Subject: [PATCH] copy --- _posts/2024-05-22-swe-bench-lite.md | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/_posts/2024-05-22-swe-bench-lite.md b/_posts/2024-05-22-swe-bench-lite.md index 0afcfc440..51e85ae87 100644 --- a/_posts/2024-05-22-swe-bench-lite.md +++ b/_posts/2024-05-22-swe-bench-lite.md @@ -20,9 +20,11 @@ The best result reported elsewhere seems to be ## Interactive, not agentic -Aider achieved this result mainly through its focus on static code analysis, +Aider achieved this result mainly through its +existing features that +focus on static code analysis, reliable LLM code editing, -and pragmatic features for AI pair programming. +and pragmatic UX for AI pair programming. Aider intentionally has quite limited and narrow "agentic behavior" to avoid long delays, high token costs and the need for users to repeatedly code review incorrect solutions.