Omarchy 4.5 Lands with Mac, ARM, Agents

// 11m agoPRODUCT LAUNCH

Omarchy 4.5 Lands with Mac, ARM, Agents

Omarchy 4.5, the agent-focused Arch Linux desktop, adds native Apple Silicon support, preliminary NVIDIA ARM support, floating workspaces with a dock, 50-language localization, and faster system performance. Its agent panel now tracks 23 providers and supports multiple Claude, Codex, and Grok accounts. Official announcement: https://omarchy.org/news/2026/11/omarchy-4-5/

// ANALYSIS

Omarchy 4.5 is less a cosmetic Linux refresh than a serious bet on the operating system as an agent workspace.

–Native M1/M2/M3/Neo support dramatically expands Omarchy beyond traditional Linux hardware, though newer Apple chips remain in development.
–The 23-provider agent panel and multi-account switching directly address the messy reality of developers juggling Claude, Codex, Cursor, Copilot, and local models.
–Floating workspaces and a dock lower the migration barrier for macOS and Windows users without abandoning Omarchy’s tiling-first identity.
–Faster installs, startup, menus, and desktop reactions make the system feel more viable as a daily development environment.
–With 740-plus pull requests from more than 180 contributors, Omarchy is becoming a cohesive agent-native platform rather than a themed Arch configuration. [GitHub release notes](https://github.com/omacom/omarchy/releases)

// TAGS

omarchyopen-sourcedevtoolagentcoding-agentai-codinglocal-first

DISCOVERED

11m ago

2026-11-05

PUBLISHED

33m ago

2026-11-05

RELEVANCE

8/ 10

AUTHOR

mark_k

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

UPDATE56m ago

OpenBot 0.60, 0.61 expand agent workflows

OpenBot’s v0.60.0 and v0.61.0 releases add visual workflow execution, Slack and Telegram triggers, multi-repository GitHub monitoring, and broader agent-provider support. Together, they push OpenBot from a local AI workspace toward an event-driven automation platform.

UPDATE2h ago

OpenBot Cuts Claude Usage 99%.

OpenBot users running Claude Code now benefit from a Claude optimization that makes background subagent status updates about 99% cheaper. Haiku 5.5 generates updates from short notes instead of rereading each subagent’s full conversation.

RESEARCH2h ago

ARC-AGI Paper Maps Training, Inference Gains

This arXiv paper evaluates training data, reasoning traces, and expert-routing calibration across three open models on ARC-AGI-2. It finds real task data and inference engineering outperform large synthetic datasets, with the best system reaching 56.67% on semi-private tasks.