0.47.0

Here's what's new:

• New models. We added support for Claude 4.5 Sonnet, Claude 4.5 Haiku, Claude Opus 4.5, GPT 5.2, Gemini 3 Pro, Gemini 3 Flash, Grok 4 and DeepSeek V3.2.
• Export chat as PDF. New Copy chat as PDF and Save chat as PDF commands let you export your conversations to PDF format.
• Resizable window. The main window can now be resized to your desired width or managed with a window management app.
• Next/prev edit navigation. Tab navigation now supports going to the next or previous edit in a conversation.
• Search improvements. Favorite chats now show on top when searching. Also, instruction name now shows for each item in the chat.
• Performance. We improved the overall app performance when you have hundreds of chats in your history.

Get on our newsletter to get notified of future updates.

0.46.2

Various bug fixes and small improvements.

0.46.1

Various bug fixes and small improvements.

0.46.0

Here's what's new:

• Better onboarding. Getting started with the app should now be easier.
• Redesigned message box. Now with a model selector and a cleaner design.
• Estimated round cost. For each round, we now show an estimated cost in cents. It doesn't take into account automatic caching that some providers do, so the actual cost is often even lower than what we show.
• Model updates. We added support for Claude 4, updated Gemini 2.5 Pro, enabled Grok to search when it needs to.
• Multiple other improvements. This update also includes a number of UI improvements, bug fixes, and performance enhancements.

Get on our newsletter to get notified of future updates.

0.45.0

New model tab management. Model tabs are now simpler and more keyboard-friendly. You can now use cmdT to open any model with the keyboard, and tabs feature helpful icons and loading states for better visual feedback.

New models. We added support for GPT 4.1, o3, Gemini 2.5 Pro, Gemini 2.5 Flash, Grok 3, and more.

Active round now follows scroll position. Previously, round-specific commands always targeted the last round. Now, as you scroll, the round currently visible in your viewport becomes the active round. This means commands will apply directly to the round you are looking at, making it easier to interact with older rounds.

Lots of other improvements. This update also includes a number of UI improvements, bug fixes, and performance enhancements.

---

Get on our newsletter to get notified of future updates. 🗞️

0.44.1

Various bug fixes and small improvements.

0.44.0

Here's what's new:

• Replaced o1-preview with o1.
• You can now pin models from the model menu — in addition to model context menu.

  

Get on our newsletter to get notified of future updates.

0.43.0

Here's what's new:

• Redesigned model tabs. The cog button is now a plus button, and selecting a model from the model menu now runs the model immediately, and now you can right-click → pin models to keep specific models in the tab bar.

  

• New models. We added o3-mini DeepSeek V3 DeepSeek R1 Gemini 2.0 Pro Exp and Gemini 2.0 Flash Thinking Exp.
• Custom models. See https://docs.intellibar.app/using-custom-models.
• License tab in Settings. Makes it easier to activate your license and see your current license status.
• The global Ask about Selection now starts a new chat. You can still use tab from the chat window to include the selection in the current chat.
• Send to Perplexity command. Similar to the Send to Google command, you can now send your prompt to Perplexity from the command palette (⌘P).
• Design refresh. More consistent spacing and colors, reorganized settings, better error states, and more.

Get on our newsletter to get notified of future updates.

0.42.0

Adds support for Gemini 2.0 Flash Exp and Gemini Exp 1206 from Google as well as Llama 3.3 70B from Groq.

Get on our newsletter to get notified of future updates.

0.41.0

Here's what's new:

• You can now favorite the current chat from the command palette (⌘P).
• Copy chat as image now breaks long responses into multiple columns.
• Settings now has a new Status tab.
• Support for o1 models OpenAI and 3.2 90B from Groq.
• Multiple design tweaks and bug fixes.

Get on our newsletter to get notified of future updates.