What’s new in Hermex 1.4
Chat and voice
- Server-first neural Listen, with on-device fallback
- Full text-to-speech controls, including play/pause, seeking, playback speed, background playback, and Lock Screen controls
- Configurable dictation preferences: Server First, On-Device First, or On-Device Only
- Reasoning controls tailored to the selected model’s capabilities
- Camera capture directly from the attachment menu
Media and sharing
- Inline audio and video previews in chats
- Downloads for unsupported file types, including
.txt,.csv,.json, and.zip - Conversation exports in HTML or JSON
- Deep links to individual sessions
Sessions, tasks, and workspaces
- Browse and unarchive archived sessions
- A dedicated Scheduled Sessions section with search and View All
- In-app workspace management: add, rename, reorder, and remove registrations without deleting files
- Scheduled-task delivery options and per-task provider selection
- Server-synced controls for CLI and Claude Code session visibility
Profiles, skills, and context
- Create profiles directly in the app
- Enable or disable skills
- New provider status screen
- Cache-efficiency statistics in Insights
- Read-only Project Context in Memory
- Localized New Chat App Shortcut phrases
Native iPad support
- Universal iPad support with sidebar and detail navigation
- Improved layouts for workspaces, files, Git, Tasks, Memory, and Insights
- Keyboard shortcuts for New Chat, search, sidebar navigation, and sending messages
Hermex requires access to a self-hosted Hermes Web UI server.