v1.4.8

Summary

This release enhances Askimo's core intelligence and user experience by introducing configurable Reasoning Effort levels and expanded model support, including Anthropic prompt caching. We have also overhauled the interface with new customization options, such as font settings, theme palettes, and optimized layout spacing. Finally, these updates include improvements to system diagnostics, token tracking, and overall application robustness.

What's Changed

• Implement configurable Reasoning Effort levels and per-model support to allow for more precise control over AI response quality (#454, #443, #453).
• Enable Anthropic prompt caching and update the Docker AI model factory to use the legacy OpenAI API for improved model compatibility (#441, #460).
• Improve visibility with new token usage tracking, search functionality for projects and sessions, and a new system diagnostics dialog (#458, #456).
• Enhance the chat experience with drag-and-drop file support, updated error displays, and refined inline control spacing (#440, #447, #446).
• Add deeper UI customization support, including new font settings for code/UI and dynamic accent color/palette styles (#444, #462).
• Standardize the application layout by migrating to Spacing constants and consolidating scrollbar styles, resulting in a more consistent visual density (#448, #451, #452).
• Refine the onboarding flow with updated UI, clearer copy, and optimized screen ordering (#434, #442).
• Improve core stability with refined localization, telemetry updates, LaTeX removal from Markdown rendering, and test suite stabilization (#445, #459, #455, #464, #463).

New Contributors

• @terry-lamp made their first contribution in #440

Full Changelog: v.1.4.7...v1.4.8