-
Improved support for thinking/reasoning models:
- Added
--thinking-tokens
CLI option to control token budget for models that support thinking. - Display thinking/reasoning content from LLMs which return it.
- Enhanced handling of reasoning tags to better clean up model responses.
- Added deprecation warning for
remove_reasoning
setting, now replaced byreasoning_tag
.
- Added
-
Aider will notify you when it's completed the last request and needs your input:
- Added notifications when LLM responses are ready with
--notifications
flag. - Specify desktop notification command with
--notifications-command
.
- Added notifications when LLM responses are ready with
-
Added/improved support for many models/providers:
- Added support for QWQ 32B.
- Support for DeepSeek V3 free on OpenRouter:
--model openrouter/deepseek/deepseek-chat:free
. - Added support for Claude 3.7 Sonnet models on OpenRouter, Bedrock and Vertex AI.
- Updated default model to Claude 3.7 Sonnet on OpenRouter.
- Added support for GPT-4.5-preview model.
- Added support for Claude 3.7 Sonnet:beta on OpenRouter.
- Added support for
openrouter/anthropic/claude-3.7-sonnet
- Fixed weak_model_name patterns to match main model name patterns for some models.
-
Switch to
tree-sitter-language-pack
for tree sitter support. -
Improved error handling for EOF (Ctrl+D) in user input prompts.
-
Added helper function to ensure hex color values have a # prefix.
-
Fixed handling of Git errors when reading staged files.
-
Improved SSL verification control for model information requests.
-
Improved empty LLM response handling with clearer warning messages.
-
Fixed Git identity retrieval to respect global configuration, by Akira Komamura.
-
Offer to install dependencies for Bedrock and Vertex AI models.
-
Deprecated model shortcut args (like --4o, --opus) in favor of the --model flag.
-
Aider wrote 85% of the code in this release.