OmniFeed Search

Google Antigravity

Google's AI code editor with completions, agents and browser-based verification

Visit antigravity.google
Screenshot of Google Antigravity
Google Antigravity, from Google

Our verdict

#7 of 7 in AI code editors · For free agent access in Google's ecosystem

Pros

  • The most useful recurring free AI package in a complete editor
  • Browser-based verification for front-end work
  • Familiar VS Code-derived editing and broad platform support

Cons

  • Recent review-workflow changes have frustrated users
  • Individual interactions can be used for improvement unless you opt out
  • No local-model or unrestricted custom-provider route

Skip if: Developers who need local models or custom providers, and anyone handling confidential code who cannot accept human review before opting out.

Read the full review →

Rankings and alternatives

AI code editors scores

Evaluated Nov 12, 2026 Evidence ↓
  • AI coding quality · counts double 3

    3/5 means: Useful everyday assistance with less consistency, maturity or supporting evidence than the leaders

    Useful Gemini-led code assistance with other curated models, repository tools and browser-based verification. Current evidence supports competitive standard tasks, but its editor execution and review reliability are less settled than the leaders and no universal editor-quality consensus exists.

  • Editor quality 3

    3/5 means: A capable main editor with evident performance, reliability or tooling tradeoffs

    The full VS Code-derived Editor view supplies familiar direct editing, commands and extension support. Current reviewer and forum reports of editor and review-workflow regressions make it a less dependable main editor than its feature list suggests.

  • Integrated AI workflow · counts double 3

    3/5 means: Good completion and agent features with noticeable friction between editing modes or surfaces

    Tab completions, natural-language commands, agent artifacts and browser verification cover a broad range of work in one IDE. Its agent-manager emphasis and changing review workflow create friction for straightforward, closely supervised coding.

  • Review and change control 3

    3/5 means: Useful diffs, approvals and recovery, but material gaps in selective control, isolation or reliability

    Approval rules, sandboxing, diffs and artifact feedback provide useful controls. Recent users report batched or missing per-file review after an update, and some current review functions require different shortcuts or surfaces, so the controls are not consistently cohesive.

  • Compatibility with existing workflows 4

    4/5 means: Broad compatibility and migration with a recognizable ecosystem or platform limitation

    A cross-platform VS Code-derived IDE with familiar settings and extension support, plus separate CLI and agent-management forms. The ranking does not treat those other forms or the other-editor extensions as substitutes for stability in the IDE itself.

  • Model and agent choice 2

    2/5 means: A curated model list with no custom provider, external agent or local inference

    The full IDE's current curated list covers Gemini, Claude and GPT-OSS models with plan restrictions. It does not provide the custom-provider, external-agent or local-model choices available in the open host editors.

  • Learning resources and support 3

    3/5 means: Adequate documentation and community help, with important AI or editor questions requiring specialist knowledge

    Familiar VS Code-derived editing and current official IDE, review and model documentation help new users get started. The newer IDE and changing review workflow have less settled community guidance, as the official forum's recent review complaints illustrate.

  • Privacy and team controls 2

    2/5 means: Individual training or human review by default, with an opt-out but limited organizational control

    Enterprise Google Cloud access offers centralized administration and regional controls. Individual interactions may be used for product and machine-learning improvement and reviewed by people unless the user changes the setting.

Pricing

  • Free Free Basic weekly agent limits; unlimited Tab completions and Command requests
  • Google AI Pro $19.99 /month Higher rate limits and a flexible AI credit pool

From the vendor’s pricing page, checked Nov 12, 2026. That page is authoritative.

Evidence and sources

AI code editors scores

AI coding quality 4 quotes
  • “Antigravity gives you access to leading frontier reasoning models from Google DeepMind and partner providers. You can switch models at any time and choose a thinking level (Low, Medium, or High) to balance response speed for quick edits against deeper multi-step reasoning for complex architecture and debugging tasks.”
    antigravity.google/docs/models · checked
  • “The fully-featured, agentic IDE. Complete with the agent manager, artifacts, and a deep understanding of your codebase.”
    antigravity.google · checked
  • “Google Antigravity's Editor view offers tab autocompletion, natural language code commands, and a configurable, and context-aware configurable agent.”
    antigravity.google/product/antigravity-ide · checked
  • “Google Antigravity is a VS Code fork that ships two modes: a traditional Editor View and a Manager View.”
    nimbalyst.com/blog/antigravity-ide-review · checked
Editor quality 4 quotes
Integrated AI workflow 4 quotes
Review and change control 2 quotes
Compatibility with existing workflows 3 quotes
  • “Available on macOS, Windows, Linux, and Googlebook”
    antigravity.google · checked
  • “The fully-featured, agentic IDE. Complete with the agent manager, artifacts, and a deep understanding of your codebase.”
    antigravity.google · checked
  • “Google Antigravity is a VS Code fork that ships two modes: a traditional Editor View and a Manager View.”
    nimbalyst.com/blog/antigravity-ide-review · checked
Model and agent choice 2 quotes
  • “Antigravity gives you access to leading frontier reasoning models from Google DeepMind and partner providers. You can switch models at any time and choose a thinking level (Low, Medium, or High) to balance response speed for quick edits against deeper multi-step reasoning for complex architecture and debugging tasks.”
    antigravity.google/docs/models · checked
  • “Antigravity gives you access to leading frontier reasoning models from Google DeepMind and partner providers. You can switch models at any time and choose a thinking level (Low, Medium, or High) to balance response speed for quick edits against deeper multi-step reasoning for complex architecture and debugging tasks.”
    antigravity.google/docs/models · checked
Learning resources and support 5 quotes
Privacy and team controls 3 quotes
  • “We use Interactions to evaluate, develop, and improve Google and Alphabet research, products, services and machine learning technologies. Interactions are collected and used in accordance with this Agreement and the Privacy Policy. Google employees and contractors may access, view, review and use Interactions. Please don't submit confidential information or any data you wouldn't want a reviewer to see or Google to use to improve our products, services and machine-learning technologies. If you don't want your Interactions used in this way, navigate to settings to change your preference on how such data is used. Even when your Telemetry is toggled off, Google still uses your data to provide the service, process any feedback and help keep Antigravity safe.”
    antigravity.google/terms · checked
  • “Gemini Enterprise license: Connect with your Gemini Enterprise Standard or Gemini Enterprise Plus seat license to get access to included first-party quotas, managed overages, and centralized administrative controls across the Antigravity App, Antigravity CLI (agy), and Antigravity IDE extensions. No separate per-user Antigravity license is required.”
    antigravity.google/docs/enterprise · checked
  • “Satisfies private networking (VPC Service Controls). Regional data residency constraints apply to first-party Gemini models; third-party model inference runs in global, us, and eu locations (in-country regions are not supported; see Anthropic's Supported countries and regions).”
    antigravity.google/docs/enterprise · checked

Facts

License 2 quotes
  • “No open-source license. The CLI, IDE and 2.0 are under Google's terms.”
    contenox.com/compare/antigravity · checked
  • “The following documents govern your use of the Service: a. the Google Terms of Service (the “Universal Terms”); b. these Google Antigravity Additional Terms of Service (the “Google Antigravity Terms”); and c. the Google Privacy Policy (the “Privacy Policy”).”
    antigravity.google/terms · checked
Runs on 2 quotes
  • “Available on macOS, Windows, Linux, and Googlebook”
    antigravity.google · checked
  • “macOS and Linux. Run curl -fsSL https://antigravity.google/cli/install.sh | bash Windows PowerShell. Run irm https://antigravity.google/cli/install.ps1 | iex Windows Command Prompt. Run curl -fsSL https://antigravity.google/cli/install.cmd -o install.cmd && install.cmd && del install.cmd”
    srithreepo, Gemini CLI maintainer, in the official transition announcement · github.com/google-gemini/gemini-cli/discussions/28017 · checked

Other quotes

Not cited by a score or fact 13 quotes
  • “Today, June 18, 2026, marks the official transition of our terminal experience from Gemini CLI to the new Antigravity CLI. Starting today, Gemini CLI will stop serving requests for Google AI Pro, Google AI Ultra, and free tier individual accounts.”
    srithreepo, Gemini CLI maintainer, in the official transition announcement · github.com/google-gemini/gemini-cli/discussions/28017 · checked
  • “For Individuals $0/month Experience Antigravity without a subscription plan”
    antigravity.google/pricing · checked
  • “Plan includes: Access to a variety of agent models Unlimited Tab completions Unlimited Command requests Basic weekly rate limits *”
    antigravity.google/pricing · checked
  • “Google AI Plus $4.99 Same as Free (Antigravity docs only give Pro and Ultra extra quota) Weekly No Google AI Pro $19.99 A high, generous quota, refreshed every five hours until the weekly limit is reached 5 hours, plus weekly cap Paid plans only, not trials Google AI Ultra 5x $99.99 The highest quota, refreshed every five hours, and the highest weekly limits (Google's subscriptions page says "higher" for 5x) 5 hours, plus weekly cap Yes Google AI Ultra 20x $199.99 Same as Ultra 5x, with "highest" rate limits per Google's subscriptions page 5 hours, plus weekly cap Yes”
    limitdigest.com/guides/antigravity-limits-explained · checked
  • “Interactive approvals: The default request-review permission asks before the agent writes files, runs shell commands, or uses web tools, unless one of your allow rules matches. In auto mode, the CLI approves workspace file edits, runs shell commands inside the terminal sandbox, and evaluates commands outside the sandbox so only risky or irreversible actions ask for your approval.”
    antigravity.google/docs/features · checked
  • “Run /diff to open the interactive diff viewer, where you can inspect workspace changes, commit history, and the changes from each agent turn, and add line-by-line comments to steer the agent's next steps.”
    antigravity.google/docs/features · checked
  • “Delegate background tasks to concurrent agent sessions. Type /agents to open the panel and monitor status, and use ctrl+k to approve tools instantly.”
    antigravity.google/product/antigravity-cli · checked
  • “Our agentic development platform Antigravity allows anyone to build in the agent-first era. It has more than 2.4 million weekly active users.”
    blog.google/company-news/inside-google/message-ceo/alphabet-earnings-q2-2026 · checked
  • “No model choice beyond the list. OpenCode, Cline and Kilo Code run any provider; Antigravity runs Google's list. No local models in the CLI. A Gemma model through Ollama works with OpenCode or Codex, not with agy. No benchmark row. Antigravity has no entry on Terminal-Bench 4.0, so it can't be ranked against Claude Code and Codex on the same tasks.”
    contenox.com/compare/antigravity · checked
  • “Run the CLI non-interactively to script agent tasks, integrate with CI pipelines, and capture machine-readable output. Pass a prompt with -p (or its aliases --print and --prompt) to run it once and exit.”
    antigravity.google/docs/features · checked
  • “Your command center to manage multiple local agents in parallel. Group conversations into Projects, operate across multiple workspaces, and automate routine tasks with scheduled messages.”
    antigravity.google · checked
  • “For Individuals $0/month Experience Antigravity without a subscription plan”
    antigravity.google/pricing · checked
  • “Plan includes: Access to a variety of agent models Unlimited Tab completions Unlimited Command requests Basic weekly rate limits *”
    antigravity.google/pricing · checked
Show all 13

Pricing

antigravity.google/pricing · checked