TL;DR

Google announced that Gemini CLI will stop working on June 18, 2026, as part of a transition to Antigravity CLI, a new platform designed for multi-agent tasks. Existing users are encouraged to migrate before the shutdown.

Google has announced that Gemini CLI will cease functioning on June 18, 2026, and will be replaced by Antigravity CLI, a new platform designed to support multi-agent workflows. This change impacts users relying on Gemini CLI for AI-assisted development and cloud provisioning, marking a significant shift in Google’s developer tools ecosystem.

The shutdown of Gemini CLI is scheduled for June 18, 2026. Google states that the transition is part of a broader effort to unify its developer tools under the Antigravity platform, which includes a server-side harness and a new terminal experience. Existing Gemini CLI features such as Agent Skills, Hooks, Subagents, and Extensions will be integrated into Antigravity CLI, although there may be some differences in feature parity initially.

Google emphasizes that Antigravity CLI offers improved performance, with faster execution thanks to its Go-based design, and supports asynchronous workflows for managing complex, multi-agent tasks. The new CLI shares architecture with Antigravity 2.0, the upcoming desktop application, ensuring seamless updates across platforms. Users are encouraged to transition early, with documentation and upcoming video tutorials to assist migration.

Why It Matters

This development is significant because it reflects Google’s strategic shift toward a unified, agent-centric development environment. For developers, it means moving from Gemini CLI, which has been a popular tool for quick AI-assisted tasks, to a more robust platform capable of handling complex, multi-agent workflows. The transition may impact workflows that depend on Gemini’s specific features, but promises a more integrated future for AI development tools.

For enterprise users, the change is less disruptive, as access to Gemini CLI remains available through paid plans, and existing licenses will continue to be supported. The move underscores Google’s focus on building an ecosystem where AI agents can communicate and operate more efficiently within a single platform.

Mastering Claude Code & Opus 4.7: From First Install to Production Workflows — The Complete CLI Guide (Claude for Developers)

Mastering Claude Code & Opus 4.7: From First Install to Production Workflows — The Complete CLI Guide (Claude for Developers)

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Background

Gemini CLI was launched in 2025 as a tool to bring AI-powered assistance directly into the terminal, gaining rapid popularity among developers. Over the past year, Google has iterated on the platform, but user feedback indicated a need for more sophisticated multi-agent capabilities and integration with other tools. The announcement aligns with Google’s broader strategy to unify its AI and developer tools into the Antigravity platform, announced earlier this year, which aims to support complex workflows and agent orchestration across multiple environments.

“We are transitioning from Gemini CLI to Antigravity CLI to better serve the evolving needs of developers working with multi-agent systems.”

— Google Developer Team

“Existing Gemini CLI users can continue to access features through paid plans, and we encourage early migration to ensure a smooth transition.”

— Google Cloud spokesperson

Design Multi-Agent AI Systems Using MCP and A2A: Engineer your own Python-based agentic AI framework with tool use, memory, and multi-agent workflows

Design Multi-Agent AI Systems Using MCP and A2A: Engineer your own Python-based agentic AI framework with tool use, memory, and multi-agent workflows

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

What Remains Unclear

It is not yet clear how the initial feature set of Antigravity CLI will compare to Gemini CLI, or what specific limitations or changes users might encounter during the transition. Details about the full feature parity and migration process are still emerging.

STM32 Debugger Programming Downloader DAPLINK Open Source Replacement ARM Cortex M Core USB Plug in 5V Portable STM32 Burner Emulator for Engineers Developers

[HIGH SPEED BURNING & DEBUGGING] This DAPLINK PRO downloader is engineered for lightning fast firmware burning and debugging…

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

What’s Next

Google will release comprehensive documentation and tutorial videos in the coming weeks to facilitate migration. Users are encouraged to start testing Antigravity CLI early to adapt workflows before the June 2026 shutdown. Further updates on feature development and platform integration are expected as the transition progresses.

Agentic Coding with Claude Code: The everyday developer's guide to agentic coding with Claude Code

Agentic Coding with Claude Code: The everyday developer's guide to agentic coding with Claude Code

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

Will my current Gemini CLI scripts stop working on June 18, 2026?

Yes, Gemini CLI will cease functioning on that date, so users should migrate to Antigravity CLI beforehand to ensure continuity.

Is there a way to access Gemini CLI after the shutdown?

Existing Gemini CLI access will be available through paid plans until the official end date, but no new requests will be processed after June 18, 2026.

What are the main differences between Gemini CLI and Antigravity CLI?

Antigravity CLI offers faster performance, support for asynchronous workflows, and a unified architecture with the Antigravity desktop application, though initial feature parity may vary.

Will enterprise users be affected by this change?

Enterprise users with existing licenses will continue to have access and support, and can use Antigravity CLI now via Google Cloud projects.

Source: Hacker News

You May Also Like

I designed a nibble-oriented CPU in Verilog to build a scientific calculator

A developer has designed a custom CPU in Verilog to create a hardware-based scientific calculator, integrating microcode firmware and FPGA tools.

What Synthetic Data Means for Model Development

No matter your goal, understanding what synthetic data means for model development can unlock new possibilities you won’t want to miss.

Medicare’s new payment model is built for AI. Most of the tech world has no idea

Medicare’s new program, ACCESS, introduces a payment model that rewards AI use for chronic care, but most of the tech world remains unaware of its significance.

Edge AI: When Intelligence Moves Closer to Devices

Persistent edge AI brings smarter devices closer to you, offering faster, more private insights—discover how it transforms everyday technology.