1.2.5

• Resolved issue #6 by providing the setting option to specify model tags

1.2.4

• Added support for GPT-4o (thanks to @ch99q)

1.2.3

• Made the chat box auto-focus after the assistant's response.
• Fixed issue where two instances of the chat view would launch on startup.

1.2.2

• Added auto-focus on chat box after assistant finishes response.

1.2.1

• Fixed bug where chat messages after the first one results in empty response.

1.2.0

• Added support for OpenRouter and Ollama as providers.
• Added feature to use few-shot prompting for inline completions and chat view (beta).
• Added feature to skip inline completions by filename and tags.
• Added settings migration system.

1.1.0

• Added a ribbon action to toggle inline completions.
• Added commands to toggle inline completions, chat view and cache ¹.

1. The commands to enable/disable these features (regardless of the current status) were already available in previous versions. This adds a new set of commands to toggle them. ↩

1.0.9

• Made "Tab" the default key for accepting inline completions (it was previously "Enter").
• Fixed bug where many of inline completions options became stale after updating the settings.

1.0.8

• Fixed a bug where stale API key is used for fetching chat and completions even after updating it in settings tab.

1.0.7

• Fixed a bug where the plugin skips fetching inline completions when cache is disabled.
• Cleaned up redundant dependencies.