HISTORY.md
/ok as a shortcut for asking aider to go ahead with the proposed changes, with optional extra instructions.xai/grok-4 and openrouter/x-ai/grok-4 model names.gemini/gemini-2.5-flash-lite-preview-06-17 model, by Tamir Zahavi-Brunner./clear now prints “All chat history cleared.” so you know it worked, by Zexin Yuan./undo output now shows only the first line of each commit message, making it easier to read.openrouter/moonshotai/kimi-k2 model, by Jack Harrington./model command.gemini-2.5-pro, gemini-2.5-flash, and gemini-2.5-pro-preview-06-05 with thinking tokens support.flash now points to gemini-2.5-flash and gemini now points to gemini-2.5-pro.--add-gitignore-files flag to enable adding files listed in .gitignore to Aider's editing scope, by omarcinkonis.--commit-language option to specify the language for commit messages, by Kyosuke Takayama./read-only command to resolve literal paths correctly, by Matteo Landi.--skip-sanity-check-repo is enabled for better performance, by Makar Ivashko.--analytics-posthog-host and --analytics-posthog-project-api-key flags, by Vasil Markoukin.claude-sonnet-4-20250514,
claude-opus-4-20250514) across various providers. The default sonnet and opus aliases were updated to these newer
versions.vertex_ai/gemini-2.5-flash-preview-05-20 model.deepseek/deepseek-r1:free for the free tier and
anthropic/claude-sonnet-4 for paid tiers.--edit-format/--editor-edit-format options./settings command now displays detailed metadata for active main, editor, and weak models..gitignore.en-US to en) and enhancing the validation of locale results.gemini-2.5-pro-preview-05-06 models.qwen3-235b models.--attribute-co-authored-by option to add co-author trailer to commit messages, by Andrew Grigorev.gemini, gemini-2.5-pro) to point to the 05-06 preview versions.overeager by default.gemini/gemini-2.5-flash-preview-04-17.gemini-2.5-pro-exp-03-25 model settings to reflect its lack of support for thinking_budget.thinking_tokens and reasoning_effort parameters for OpenRouter models.litellm.completion_cost where available.openrouter/google/gemini-2.5-pro-preview-03-25.--disable-playwright flag to prevent Playwright installation prompts and usage, by Andrew Grigorev.aider scrape command-line tool will now use Playwright for web scraping if it is available, by Jon Keys.oslex for argument quoting, by Titusz Pan.oslex for robust argument quoting, by Titusz Pan./ask mode to instruct the LLM to elide unchanging code in its responses.--disable-playwright flag.reasoning_effort for Gemini 2.5 Flash models.--shell-completions argument to generate shell completion scripts (e.g., for bash, zsh).--attribute-author or --attribute-committer flags now override the default behavior when --attribute-co-authored-by is used, allowing finer control over commit attribution, by Andrew Grigorev./drop after adding a read-only file).aider-args utility (or python -m aider.args) now defaults to printing a sample YAML configuration if no arguments are provided.gemini-2.5-flash-preview-04-17 models.udiff-simple edit format, for Gemini 2.5 Pro.gemini-2.5-flash-preview-04-17.--- /dev/null)./commit, by Kay Gosho.overeager prompt setting by default..svg, .pdf) to the default list of ignored files for AI comment scanning (--watch).--watch).o3 and o4-mini including provider-specific versions for OpenAI, OpenRouter, and Azure.gpt-4.1 and gpt-4.1-mini models.o3 models since you need identity verification to stream.patch edit format for OpenAI's GPT-4.1 model.editor-diff, editor-whole, and editor-diff-fenced edit formats.grok-3-fast-beta and grok-3-mini-fast-beta models.xai/grok-3-beta, xai/grok-3-mini-beta, openrouter/x-ai/grok-3-beta, openrouter/x-ai/grok-3-mini-beta, and openrouter/openrouter/optimus-alpha models.xai/grok-3-beta.openrouter/openrouter/optimus-alpha.aider.conf.yml.gemini/gemini-2.5-pro-preview-03-25 model.gemini alias to point to gemini/gemini-2.5-pro-preview-03-25.gemini-exp alias for gemini/gemini-2.5-pro-exp-03-25.openrouter/openrouter/quasar-alpha model.
aider --model quasargemini/gemini-2.0-flash and openrouter/google/gemini-2.0-flash-exp:free) for Gemini 2.5 Pro models.openrouter/google/gemini-2.0-flash-exp:free.OPENROUTER_API_KEY is set and no model is specified.gemini/gemini-2.5-pro-exp-03-25 if GEMINI_API_KEY is set, and vertex_ai/gemini-2.5-pro-exp-03-25 if VERTEXAI_PROJECT is set, when no model is specified.--stream and --cache-prompts are used together, as cost estimates may be inaccurate./model, if the user was using the old model's default format.Ctrl-X Ctrl-E keybinding to edit the current input buffer in an external editor, by Matteo Landi.openrouter/deepseek-chat-v3-0324:free model./run that was preventing auto-testing.UnboundLocalError during git tree traversal.GitCommandNotFound error if git is not installed or not in PATH.FileNotFoundError if the current working directory is deleted while aider is running./context command that automatically identifies which files need to be edited for a given request./edit as an alias for the /editor command./editor-model for Editor Model, and /weak-model for Weak Model, by csala.--reasoning-effort and --thinking-tokens if the model doesn't support them.--check-model-accepts-settings flag (default: true) to force unsupported model settings.--git-commit-verify flag (default: False) to control whether git commit hooks are bypassed./ask, /code, and /architect commands, by shladnik.--aiderignore argument to resolve both absolute and relative paths, by mopemope.openrouter/google/gemma-3-27b-it model./think-tokens command to set thinking token budget with support for human-readable formats (8k, 10.5k, 0.5M)./reasoning-effort command to control model reasoning level./think-tokens and /reasoning-effort commands display current settings when called without arguments.--thinking-tokens argument to accept string values with human-readable formats.--auto-accept-architect flag (default: true) to automatically accept changes from architect coder format without confirmation.cohere_chat/command-a-03-2025 and gemini/gemma-3-27b-it/drop command now preserves original read-only files provided via args.read.--shortcut switches even when already specified in the command line.--thinking-tokens CLI option to control token budget for models that support thinking.remove_reasoning setting, now replaced by reasoning_tag.--notifications flag.--notifications-command.tree-sitter-language-pack for tree sitter support.--model openrouter/deepseek/deepseek-chat:free.openrouter/anthropic/claude-3.7-sonnet--model sonnet to use the new 3.7/editor command.<think> tags from R1 responses for commit messages (and other weak model uses).use_temperature: <float> in model settings, not just true/false.boto3 for Bedrock.HOME=/app which is the normal project mount-point, to persist ~/.aider.python, php, etc.--timeout/model now correctly reports that the weak model is not changed.node_modules.aider --model o3-mini--reasoning-effort argument: low, medium, high.remove_reasoning: tagname model setting.--model openrouter/deepseek/deepseek-r1:freeopenrouter/deepseek/deepseek-r1--model r1--model openrouter/deepseek/deepseek-r1--line-endings for file writing, by Titusz Pan./ask and /code./model o1 and a streaming model/ask, /code and /architect commands now switch the chat mode.--subtree-only, and only watches that subtree./load or --load.# ... AI? comments to trigger aider and ask questions about your code.# AI comments, // AI comments, or -- AI comments to give aider instructions in any text file.aider --model flash or aider --model gemini/gemini-2.0-flash-exp--multiline flag and /multiline-mode command makes ENTER a soft newline and META-ENTER send the message, by @miradnanali./copy-context <instructions> now takes optional "instructions" when copying code context to the clipboard./voice now lets you edit the transcribed text before sending.--copy-paste mode./copy-context command.--api-key provider=key setting.--set-env VAR=value setting.--watch-files.sys.python for self-upgrades.aider --watch-files and it will watch for instructions you add to your source files.# ... or // ... comments that start or end with "AI" are instructions to aider./run or /test have non-zero exit codes, pre-fill "Fix that" into the next message prompt./diff now invokes git diff to use your preferred diff tool.--read now expands ~ home dirs.--voice-input-device to select audio input device for voice recording, by @preynal.--timeout option to configure API call timeouts./drop command to support substring matching for non-glob patterns.--alias.--alias config to define custom model aliases.--[no-]detect-urls flag to disable detecting and offering to scrape URLs found in the chat./editor returns and prefill the file content into the prompt, so you can use /editor to compose messages that start with /commands, etc./editor command to open system editor for writing prompts, by @thehunmonkgroup.gpt-4o-2024-11-20./run and suggested shell commands are less mysterious and now confirm that they "Added XX lines of output to the chat."extra_params for litellm.completion().--show-release-notes to control release notes display on first run of new version.--read./web command just adds the page to the chat, without triggering an LLM response.--haiku to use it.aider --apply-clipboard-edits file-to-edit.js./save <fname> command will make a file of /add and /read-only commands that recreate the current file context in the chat./load <fname> will replay the commands in the file./load to run any arbitrary set of slash-commands, not just /add and /read-only.--load <fname> to run a list of commands on launch, before the interactive chat begins.supports_vision attribute to enable image support for models./add and /read-only.--no-fancy-input switch disables prompt toolkit input, now still available with --no-pretty.--no-browser or --no-gui..env to the .gitignore file./help RAG results.yes: true in YAML config, show helpful error./read-only:
/add.src/**/*.py--yes to --yes-always.
AIDER_YES_ALWAYS env var and yes-always: YAML key.--yes on the command line. - list entries, one per line./settings now includes the same announcement lines that would print at launch.--editor-model on launch now, same as main and weak models.--skip-sanity-check-repo switch to speedup launch in large repos.--o1-preview and --o1-mini shortcuts./read-only supports the entire filesystem./copy command to copy the last LLM response to the clipboard./clipboard to /paste.--voice-format switch to send voice audio as wav/mp3/webm, by @mbailey.extra_params dict to specify any extras to pass to litellm.completion().aider --model o1-miniaider --model o1-preview/run correctly uses PowerShell or cmd.exe./read-only.input() if --no-pretty or a Windows console is not available./report command to open your browser with a pre-populated GitHub Issue.--chat-language switch to set the spoken language.--[no-]suggest-shell-commands controls both prompting for and offering to execute shell commands.--models to --list-models by @fry69.--tool-warning-color setting./command./add, errors writing files./voice transcription exceptions, show the WAV file so the user can recover it.vX.Y.Z.dev version naming./settings command to show active settings.--verbose.--update as an alias for --upgrade./model command.watchdog<5 for docker image.pexpect dependency.gemini/gemini-1.5-pro-exp-0827 and gemini/gemini-1.5-flash-exp-0827./run commands can now be interactive in environments where a pty is available.--[no-]suggest-shell-commands switch to configure shell commands.--upgrade switch to install latest version of aider from pypi.--show-prompt./undo for all models./web output.--cache-keepalive-pings.
extra_headers, to support Anthropic beta features.--install-main-branch to update to the latest dev version of aider.--verbose.\_./read and /drop now expand ~ to the home dir./reset command to /drop files and /clear chat history.--map-multiplier-no-files to control repo map size multiplier when no files are in the chat.
gpt-4o-2024-08-06.--cache-prompts.
/read-only files.--map-refresh <always|files|manual|auto> to configure..ipynb files.--verbose..aider.models.metadata.json data.--msg /ask ... caused an exception.--deepseek switch to use DeepSeek Coder.--chat-mode <mode> switch to launch in ask/help/code modes./code <message> command request a code edit while in ask mode./read only files.setup.py to pyproject.toml, by @branchvincent./ask./tokens./help./read and --read, including from outside the git repo./diff now shows diffs of all changes resulting from your request, including lint and test fixes./clipboard command to paste images or text from the clipboard, replaces /add-clipboard-image./web./ commands.--mini for gpt-4o-mini./ask and /help./tokens now shows the active model.--attribute-commit-message-author to prefix commit messages with 'aider: ' if aider authored the changes, replaces --attribute-commit-message.--attribute-commit-message-committer to prefix all commit messages with 'aider: '.openai/gpt-4o-2024-08-06.extra_headers.--subtree-only to limit aider to current directory subtree.
/add-clipboard-image to add images to the chat from your clipboard.--map-tokens 1024 to use repo map with any model.--commit-prompt to customize the commit message prompt.--lint on all dirty files in repo./run).docker run --user xxx.paulgauthier/aider-full docker image, which includes all extras./commands without a completion override./run add to chat approval logic./ask <question> command to ask about your code, without making any edits./chat-mode <mode> command to switch chat modes:
file: CONVENTIONS.md to .aider.conf.yml to always load a specific file.
file: [file1, file2, file3] to always load multiple files./add and /drop is now case-insensitive.--no-verify-ssl:
/web scrapes websites./web scraping functionality/undo with new files or new repos.--verbose output..env files.aider-chat[help]aider-chat[browser]aider-chat[playwright]/add.--models.--check-update switch was renamed to --just-check-updated.--skip-check-update switch was renamed to --[no-]check-update./help <question> to ask for help about using aider, customizing settings, troubleshooting, using LLMs, etc./undo.$HOME/.aider/caches dir for app-wide expendable caches.--model-settings-file is now .aider.model.settings.yml.--model-metadata-file is now .aider.model.metadata.json.--no-git.claude-3-5-sonnet-20240620 by default if ANTHROPIC_API_KEY is set in the environment.--attribute-commit-message to prefix aider's commit messages with "aider:".overloaded_error./undo so it works regardless of --attribute settings.--help functionality.--model-settings-file to customize per-model settings, like use of repo-map (by @caseymcc).--sonnet for Claude 3.5 Sonnet, which is the top model on aider's LLM code editing leaderboard.AIDER_xxx environment variables can now be set in .env (by @jpshack-at-palomar).--llm-history-file to log raw messages sent to the LLM (by @daniel-vainsencher).--vim for vim keybindings in the chat..aider.models.json file (by @caseymcc).--commit --dry-run does nothing.make or run which may just be words.GIT_EDITOR env for commits if it is already set./voice./lint in the chat or --lint on the command line.--lint-cmd.--test-cmd./test or from the command line with --test.--restore-chat-history to restore prior chat history on launch, so you can continue the last conversation.httpx errors.openrouter/anthropic/claude-3-opus, gpt-3.5-turbo--show-prompts debug switch.DEEPSEEK_API_KEY and deepseek/deepseek-chat, etc rather than as a generic OpenAI compatible API.--browser switch to launch an experimental browser based version of aider./model <name> and search the list of available models with /models <query>.google-generativeai dependency--models <MODEL-NAME> to search the available models.--no-show-model-warnings to silence model warnings.--weak-model <model-name> which allows you to specify which model to use for commit messages and chat history summarization.--4-turbo-vision--opus--sonnet--anthropic-api-key/add of images from anywhere in the filesystem.gpt-4-turbo-2024-04-09 and gpt-4-turbo models.
gpt-4-0613 and worse than the gpt-4-preview-XXXX models. See recent Exercism benchmark results.gpt-4-preview-XXXX models. See recent refactor bencmark results.gpt-4-1106-preview as it performs best on both benchmarks, and significantly better on the refactoring/laziness benchmark.-3 and -4 switches to use GPT 3.5 or GPT-4 (non-Turbo)..aiderignore
--openai-organization-id argument./web <url> command which scrapes the url, turns it into fairly clean markdown and adds it to the chat.gpt-3.5-turbo-0125.! alias for /run.--model gpt-4-0125-preview and OpenAI's alias --model gpt-4-turbo-preview. The --4turbo switch remains an alias for --model gpt-4-1106-preview at this time./test command that runs a command and adds the output to the chat on non-zero exit status./quit as alias for /exit.--skip-check-update to skip checking for the update on launch.--openrouter as a shortcut for --openai-api-base https://openrouter.ai/api/v1OPENAI_API_BASE, OPENAI_API_TYPE, OPENAI_API_VERSION, OPENAI_API_DEPLOYMENT_ID.--check-update arg to check if updates are available and exit with status code.Add images to the chat to automatically use GPT-4 Vision, by @joshuavial
Bugfixes:
/run command output, by @ctothgpt-4-1106-preview.--4-turbo command line switch as a shortcut for --model gpt-4-1106-preview.gpt-4-1106-preview).gpt-4-0613). Performance leaps from 47%/64% up to 51%/71%./add used on file matching .gitignore/add wildcards..aiderignore file, which instructs aider to ignore parts of the git repo.--commit cmd line arg, which just commits all pending changes with a sensible commit message generated by gpt-3.5./run and /git now accept full shell commands, like: /run (cd subdir; ls)--encoding cmd line switch.soundfile package not available.--voice-language switch to constrain /voice to transcribe to a specific languagesounddevice, as it was slowing down aider startuptext when using --no-pretty.--no-stream/git command to run git from inside aider chats..gitignore with .aider* to prevent users from accidentally adding aider files to git.--no-pretty if aider detects it is running inside a VSCode terminal.--show-repo-map--verbose output--dark-mode and --light-mode to select colors optimized for terminal background--help output/add and /drop commands--encoding into ctags, require it to return utf-8--encoding switch to specify another encoding--code-theme switch to control the pygments styling of code blocks (by @kwmiebach)--message to give GPT that one instruction and then exit after it replies and any edits are performed.--no-stream to disable streaming GPT responses.
gpt-3.5-turbo-16k, and all OpenAI chat models--openai-api-base to support API proxies, etc.gpt-3.5-turbo and gpt-4-32k.--map-tokens to set a token budget for the repo map, along with a PageRank based algorithm for prioritizing which files and identifiers to include in the map./tokens to report on context window token usage./clear to clear the conversation history.