# aider.chat

> AI-optimized mirror of aider.chat containing 50 pages totalling 39,379 words of clean markdown content, structured data, and semantic HTML. Original source: https://aider.chat/. Last updated: 2026-05-01T14:23:31.160Z.

Each page is available as HTML (with JSON-LD structured data) and Markdown (text-only, ideal for LLMs and RAG).

## Homepage

- [AI pair programming in your terminal](/site-root.html) (475 words)

## Articles & Blog Posts

- [2024/06/02/main-swe-bench-html.html](/2024/06/02/main-swe-bench-html.html) (1 words)
- [2023/12/21/unified-diffs-html.html](/2023/12/21/unified-diffs-html.html) (1 words)
- [2023/07/02/benchmarks-html.html](/2023/07/02/benchmarks-html.html) (1 words)
- [2023/05/25/ctags-html.html](/2023/05/25/ctags-html.html) (1 words)
- [docs/config/options-html.html](/docs/config/options-html.html) (1 words)
- [LLMs are bad at returning code in JSON](/2024/08/14/code-in-json-html.html): LLMs write worse code if you ask them to return the code wrapped in JSON via a tool function call. (1,248 words)
- [docs/config/aider_conf-html.html](/docs/config/aider_conf-html.html) (1 words)
- [2024/05/22/swe-bench-lite-html.html](/2024/05/22/swe-bench-lite-html.html) (1 words)
- [docs/benchmarks-html.html](/docs/benchmarks-html.html) (1 words)
- [docs/usage/commands-html.html](/docs/usage/commands-html.html) (1 words)
- [FAQ](/docs/faq-html.html): Frequently asked questions about aider. (2,126 words)
- [Linting code for LLMs with tree-sitter](/2024/05/22/linting-html.html): Aider now lints code after every LLM edit and automatically fixes errors, using tree-sitter and AST-aware code context. (732 words)
- [Aider Documentation](/docs/index.html): aider is AI pair programming in your terminal (825 words)
- [Coding with Llama 3.1, new DeepSeek Coder & Mistral Large](/2024/07/25/new-models-html.html): Summary of code editing skill for the new models, with Sonnet and GPT-3.5 for scale. (377 words)
- [Building a better repository map with tree sitter](/2023/10/22/repomap-html.html): Tree-sitter allows aider to build a repo map that better summarizes large code bases. (1,577 words)
- [Alternative DeepSeek V3 providers](/2025/01/28/deepseek-down-html.html): DeepSeek’s API has been experiencing reliability issues. Here are alternative providers you can use. (800 words)
- [Claude 3 beats GPT-4 on Aider’s code editing benchmark](/2024/03/08/claude-3-html.html): Claude 3 Opus outperforms all of OpenAI’s models on Aider’s code editing benchmark, making it the best available model for pair programming with AI. (623 words)
- [Sonnet is the opposite of lazy](/2024/07/01/sonnet-not-lazy-html.html): Claude 3.5 Sonnet can easily write more good code than fits in one 4k token API response. (651 words)
- [Aider has written 7% of its own code (outdated, now 70%)](/2024/05/24/self-assembly-html.html): This article is quite out dated. Aider is currently writing about 70% of the new code in each release. (331 words)
- [Tips](/docs/usage/tips-html.html): Tips for AI pair programming with aider. (754 words)
- [Code editing benchmarks for OpenAI’s “1106” models](/2023/11/06/benchmarks-1106-html.html): A quantitative comparison of the code editing capabilities of the new GPT-3.5 and GPT-4 versions that were released in Nov 2023. (666 words)
- [Drawing graphs with aider, GPT-4o and matplotlib](/2024/05/13/models-over-time-html.html): Use GPT-4o to draw graphs with matplotlib, including adjusting styles and making visual changes. You get the graph, but you also get the code in your repo. (1,208 words)
- [Chat modes](/docs/usage/modes-html.html): Using the code, architect, ask and help chat modes. (1,001 words)
- [Linting and testing](/docs/usage/lint-test-html.html): Automatically fix linting and testing errors. (528 words)
- [GPT-4 Turbo with Vision is a step backwards for coding](/2024/04/09/gpt-4-turbo-html.html): OpenAI’s GPT-4 Turbo with Vision model scores worse on aider’s code editing benchmarks than all the previous GPT-4 models. In particular, it seems much more prone to “lazy coding” than the existing GPT-4 Turbo “preview” models. (348 words)
- [Sonnet seems as good as ever](/2024/08/26/sonnet-seems-fine-html.html): Sonnet’s score on the aider code editing benchmark has been stable since it launched. (201 words)
- [The January GPT-4 Turbo is lazier than the last version](/2024/01/25/benchmarks-0125-html.html): The new gpt-4-0125-preview model is quantiatively lazier at coding than previous GPT-4 versions, according to a new “laziness” benchmark. (226 words)
- [Scripting aider](/docs/scripting-html.html): You can script aider via the command line or python. (377 words)
- [Speed benchmarks of GPT-4 Turbo and gpt-3.5-turbo-1106](/2023/11/06/benchmarks-speed-1106-html.html): This report provides a detailed comparison of the speed of GPT-4 Turbo and gpt-3.5-turbo-1106 models based on the aider benchmarking suite. (315 words)
- [Model warnings](/docs/llms/warnings-html.html): aider is AI pair programming in your terminal (300 words)
- [Aider in your browser](/2024/05/02/browser-html.html): Aider has an experimental browser UI, allowing you to collaborate with LLMs on code in your local git repo. (135 words)
- [Installation](/docs/install-html.html): How to install and get started pair programming with aider. (452 words)
- [Repository map](/docs/repomap-html.html): Aider uses a map of your git repository to provide code context to LLMs. (586 words)
- [Git integration](/docs/git-html.html): Aider is tightly integrated with git. (549 words)
- [Specifying coding conventions](/docs/usage/conventions-html.html): Tell aider to follow your coding conventions when it works on your code. (434 words)
- [Edit formats](/docs/more/edit-formats-html.html): Aider uses various “edit formats” to let LLMs edit source files. (468 words)
- [A draft post](/2024/05/22/draft-html.html): With a draft summary. (21 words)
- [Ollama](/docs/llms/ollama-html.html): aider is AI pair programming in your terminal (260 words)
- [Usage](/docs/usage-html.html): How to use aider to pair program with AI and edit code in your local git repo. (475 words)
- [Aider can connect to most LLMs](/docs/llms-html.html): Aider can connect to most LLMs for AI pair programming. (207 words)
- [DeepSeek](/docs/llms/deepseek-html.html): aider is AI pair programming in your terminal (105 words)
- [OpenRouter](/docs/llms/openrouter-html.html): aider is AI pair programming in your terminal (266 words)
- [OpenAI compatible APIs](/docs/llms/openai-compat-html.html): aider is AI pair programming in your terminal (120 words)
- [sitemap-xml.html](/sitemap-xml.html) (162 words)
- [Images & web pages](/docs/usage/images-urls-html.html): Add images and web pages to the aider coding chat. (243 words)
- [Release history](/history-html.html): Release notes and stats on aider writing its own code. (10,216 words)
- [Aider LLM Leaderboards](/docs/leaderboards/index.html): Quantitative benchmarks of LLM code editing skill. (8,645 words)
- [Configuration](/docs/config-html.html): Information on all of aider’s settings and how to use them. (141 words)
- [Prompt caching](/docs/usage/caching-html.html): Aider supports prompt caching for cost savings and faster coding. (196 words)

## Resources

- [Full Page Index](/index.html): Browse all cached pages with rich metadata
- [About This Cache](/content/about.html): Methodology, technical details, and usage guidelines
- [XML Sitemap](/content/sitemap.xml): Machine-readable sitemap for crawler discovery
- [Robots.txt](/content/robots.txt): Crawler directives