|
9 | 9 | * **Why the checkbox?** This setting is provided as a temporary workaround for potential response delays sometimes observed with Google's caching mechanism when accessed via these two providers. Caching is *not* enabled by default for them. |
10 | 10 | * **Requesty:** Caching remains automatic for supported models via Requesty. |
11 | 11 |
|
| 12 | +## Boomerang Orchestrator Mode |
| 13 | + |
| 14 | +* Added Boomerang Orchestrator as a built-in mode. See the [Boomerang Tasks documentation](/features/boomerang-tasks) for more details. |
| 15 | + <img src="/img/v3.14.3/v3.14.3-1.png" alt="Boomerang Orchestrator Mode" width="600" /> |
| 16 | + |
| 17 | +## Sexy new minimalist look |
| 18 | + |
| 19 | +* Improved the home screen user interface for a better initial experience. |
| 20 | + <img src="/img/v3.14.3/v3.14.3.png" alt="New Minimalist Home Screen" width="600" /> |
| 21 | += |
12 | 22 | ## Apply Diff and Other Major File Edit Improvements |
13 | 23 |
|
14 | 24 | * Improve [`apply_diff`](/features/tools/apply-diff) to work better with **Google Gemini 2.5** and other models |
|
21 | 31 | * Ensure user feedback is added to conversation history even during API errors (thanks System233!). |
22 | 32 | * Prevent redundant 'TASK RESUMPTION' prompts from appearing when resuming a task (thanks System233!). |
23 | 33 | * Fix issue where error messages sometimes didn't display after cancelling an API request (thanks System233!). |
24 | | -* Preserve editor state and prevent tab unpinning during diffs (thanks seedlord!) |
25 | 34 |
|
26 | 35 | ## Terminal Fixes |
27 | 36 |
|
|
59 | 68 | * Improvements to Requesty model list fetching (thanks dtrugman!) |
60 | 69 | * Make the VS Code LM provider show the correct model information (thanks QuinsZouls!) |
61 | 70 | * Remove unnecessary calculation from VS Code LM provider (thanks d-oit!) |
| 71 | +* Updated the default model ID for the Unbound provider from `claude-3.5-sonnet` to `claude-3.7-sonnet` (thanks pugazhendhi-m!). |
62 | 72 |
|
63 | 73 | ## Bug Fixes and General QOL Improvements and Misc. |
64 | 74 |
|
65 | 75 | * Make the [`list_files`](/features/tools/list-files) tool more efficient and smarter about excluding directories like `.git/` |
66 | | -* Performance improvements to task size calculations |
| 76 | +* Performance improvements to task size calculations, including more efficient token count estimation to reduce gray screens. |
67 | 77 | * Give better loading feedback on chat rows (thanks elianiva!) |
68 | 78 | * Use a more sensible task export icon |
69 | 79 | * Fix file drag and drop on Windows and when using SSH tunnels (thanks NyxJae!) |
|
75 | 85 | * Track tool use errors in evals |
76 | 86 | * Use path aliases in webview source files |
77 | 87 | * Better handling of FakeAI “controller” object (thanks wkordalski) |
| 88 | +* Ensure user feedback is added to conversation history even during API errors (thanks System233!). |
| 89 | +* Prevent redundant 'TASK RESUMPTION' prompts from appearing when resuming a task (thanks System233!). |
| 90 | +* Fix issue where error messages sometimes didn't display after cancelling an API request (thanks System233!). |
| 91 | +* Preserve editor state and prevent tab unpinning during diffs (thanks seedlord!) |
| 92 | +* Cleaned up the internal settings data model for better maintainability. |
| 93 | +* Optimized API calls by omitting reasoning parameters for models that do not support reasoning. |
| 94 | +* Corrected word wrapping in Roo message titles (thanks zhangtony239!). |
| 95 | +* Improved clarity in the documentation regarding adding custom settings (thanks shariqriazz!). |
0 commit comments