Back to vendors
G

Google Antigravity

Also known as: Antigravity, Antigravity 2.0, AGY, Antigravity CLI, Antigravity SDK, Gemini CLI, Gemini Code Assist

Visit site
Entry priceFree for individuals. Google AI Pro is $19.99/mo, and Ultra starts at $99.99/mo.Full pricing detail

Google's agentic development platform across a desktop app, CLI, Python SDK and IDE extensions, where agents work autonomously across editor, terminal and browser with model choice spanning Gemini and Claude.

Google Antigravity is an agentic development platform spanning four surfaces: a standalone desktop app for managing agents in parallel, a terminal CLI including a headless mode, a Python SDK for building custom agents on the Antigravity harness, and IDE extensions for VS Code, Visual Studio, JetBrains, Zed and Xcode. It launched as a public preview, has reached version 2.0 and absorbs the retired Gemini CLI.

In the Manager surface, users spawn, orchestrate and observe multiple agents working asynchronously across workspaces rather than supervising individual tool calls. Agents operate autonomously across the editor, terminal and a Chrome browser. A single task can write a feature, launch the application from the terminal, then drive the browser to verify it works. Asynchronous subagents let the main agent delegate parallel background work, and a Teamwork preview coordinates agent teams.

Visual Artifacts turn the agent's work into reviewable reports, among them implementation plans, task plans, code diffs, walkthroughs, screenshots and browser recordings. Users can highlight any section and leave comments to steer the agent, much like commenting in a document. Permissions ship with safe defaults, local proxying and granular tool approval gates, and browser access is bounded by allowlist, denylist and a separate Chrome profile.

Developers choose among Google's Gemini models, led by Gemini 4 Argon on paid plans and Gemini 3.8 Flash on the free tier, and on paid plans Anthropic's Claude Sonnet 5.5 and Opus 5.5. The SDK can also run agents on local models. Customization runs through MCP, Skills, Rules, Workflows, Plugins, Hooks and Sidecars, with Google publishing curated bundles for Android, Firebase, Web and DeepMind science work. Enterprise deployment runs models inside the customer's own Google Cloud project under Gemini Enterprise, with selectable global, US or EU endpoints. It adds VPC Service Controls, Workforce Identity Federation for bring your own identity, and request and response logging.

Vendor details

Canonical URL

https://antigravity.google

Category

Coding agent

Subcategory

Coding — agentic IDE

Funding status

Google product, not an independently funded company. In public preview since November 2025, now at Antigravity 2.0 (v2.11.0) with CLI v1.1.22, SDK v0.1.15 and IDE extensions v2.5.5. Absorbs the retired Gemini CLI and offers a Firebase Studio migration path.

Company status

first party product

Use cases & customers

Primary use cases

agentic codingsoftware development

Target customers

developers

Deployment options

Desktop app (macOS, Linux, Windows)CLI (including headless)Python SDKIDE extensions (VS Code, Visual Studio, JetBrains, Zed, Xcode)Gemini Enterprise on customer Google Cloud project (global, us or eu endpoints, VPC-SC)

Integrations

MCP client support across the desktop app, CLI, SDK and IDE, with Google published Chrome and Web MCP servers for autonomous browser research. The agent operates the editor, terminal and a Chrome browser as tools, and Plugins, Hooks and Sidecars extend behavior. IDE extensions cover VS Code, Visual Studio, JetBrains, Zed and Xcode, and curated Google bundles ship for Android, Firebase, Web and DeepMind science workflows.

In practice

You want an agentic coding tool that works in the editor, the terminal, and your own scripts. Google Antigravity spans desktop, CLI, and an SDK, so the agent meets your workflow.

An agent that can't see the running app guesses at front-end work. Antigravity has a built-in browser, so it can observe and act on what it builds.

You don't want to be tied to a single model in your IDE. Antigravity supports multiple models and MCP, so you connect the tools and models you choose.

Agentic Index coverage score

12.5 / 14 capabilities · 89%

Integrations & Tool Calling Full

MCP tools connect the agent to outside systems under per tool allow, ask or deny rules, plugins bundle MCP servers and skills managed from the Customizations tab, and the agent operates the editor, terminal and a browser as tools.

Sourceantigravity.google/docs/permissions, docs/changelog and docs/sdk/overviewread 2026-09-29

Workflow Orchestration Full

The main agent delegates to custom subagents, whose invocations appear as cards with live status, custom agents can switch off default prompt sections and tools, a Teamwork preview coordinates agent teams, and the SDK builds multi agent systems with subagents.

Sourceantigravity.google/docs/changelog, docs/sdk/overview and llms.txt; antigravity.google/docs/subagentsread 2026-09-29

Knowledge Grounding & RAG Partial

The agent reads the project's code with codebase search, rules (now with their own token budget) and installed skill bundles, and reaches outside data through MCP. Context is gathered on each run, and Google does not describe a persistent index, graph or embeddings layer over the customer's code.

Sourceantigravity.google/docs/build-with-google, docs/changelog and llms.txtread 2026-09-29

Human Oversight & Guardrails Full

Permission lists with Deny over Ask over Allow make the agent pause for explicit approval before matching commands, file access, URL navigation or MCP tool use, Request Review mode confirms every command, web browsing and MCP tools prompt by default, administrators can enforce enterprise policies, and /plan lets users read and edit the plan before any code is written.

Sourceantigravity.google/docs/permissions and docs/changelogread 2026-09-29

Security, Identity & Governance Full

Enterprise use runs through Gemini Enterprise in the customer's Google Cloud project with single sign on, Workforce Identity Federation for providers such as Okta, Application Default Credentials for headless use, VPC Service Controls, request and response logging and enforced enterprise permission policies, under Google Cloud's enterprise security controls and terms. Google does not name an attestation specific to Antigravity or say which Google Cloud attestations cover it.

Sourceantigravity.google/docs/enterprise and docs/permissionsread 2026-09-29

Observability & Auditability Full

The agent records its work as reviewable artifacts (implementation plans, task lists, walkthroughs, screenshots and browser recordings), subagent invocations show as cards with live status, and enterprise deployments log requests and responses for audit trails.

Sourceantigravity.google/docs/changelog, docs/enterprise and llms.txt (Artifacts, Walkthrough, Screenshots, Browser Recordings); antigravity.google/docs/artifactsread 2026-09-29

Memory & State Persistence Partial

Projects group conversations across workspaces and worktrees, and conversations resume with /resume, which carries session history. Google does not describe a store the agent writes and reads across separate sessions, rules are standing instructions, and cron sidecars schedule work.

Sourceantigravity.google llms.txt (Projects, Rules, Sidecars) and docs/sidecarsread 2026-09-29

Deployment & Data Residency Full

Enterprise deployment runs inside the organization's own Google Cloud project with global, US multi region or EU multi region endpoints for data residency and VPC Service Controls, and Google states enterprise prompts, responses, code and telemetry are never stored outside the customer's private environments.

Sourceantigravity.google/docs/enterpriseread 2026-09-29

Prebuilt Agents, Templates & Packs Full

Google publishes named bundles, among them Modern Web Guidance, a Firebase bundle, Android CLI, DeepMind Science skills, Chrome DevTools, Dart and Flutter, a Data Agent Kit starter pack and Google Maps Platform, and plugins are discovered, installed and managed from the Customizations tab.

Sourceantigravity.google/docs/build-with-google and docs/changelogread 2026-09-29

Triggers & Channel Coverage Full

A sidecar's schedule builtin runs agent commands, such as starting a new conversation, on a standard cron expression. Agents are also invoked from the desktop app, the CLI including headless mode, the Python SDK and IDE extensions, and by remote control.

Sourceantigravity.google/docs/sidecars and llms.txtread 2026-09-29

Model Flexibility & Routing Full

Users pick the reasoning model from a selector. It spans Google's Gemini models (Gemini 4 Argon on paid plans, Gemini 3.8 Flash on the free tier) and, on paid Pro, Ultra and Enterprise, Anthropic's Claude Sonnet 5.5 and Opus 5.5. The Claude 4.6 models and gpt-oss-120b are listed for removal on 2 November 2026. The SDK runs agents on device with LiteRT or local OpenAI compatible servers, so customers can also bring their own model.

Sourceantigravity.google/docs/models and docs/sdk/overviewread 2026-10-25

APIs, SDKs & MCP Extensibility Full

The Antigravity SDK is a Python SDK for building autonomous agents on the Antigravity harness with custom tools, safety policies, subagents, lifecycle hooks, structured output and local models, beside MCP support, plugins, hooks and sidecars.

Sourceantigravity.google/docs/sdk/overview and llms.txtread 2026-09-29

Testing, Debugging & Optimization Partial

The agent verifies its own work by running the app in a browser and producing screenshots, recordings and walkthroughs, and a terminal sandbox bounds commands. Google does not describe a harness, scored test set or optimization loop for evaluating agent behavior, and an evaluation preset reported for SDK 0.1.18 does not appear on its SDK overview or feature pages.

Sourceantigravity.google/docs/sdk/overview and llms.txtread 2026-09-29

Browser & Computer Use Full

The agent drives a Chrome browser itself to open, click, type in and verify the app it builds, with URL navigation and browsing prompting for approval by default, and captures screenshots and browser recordings as artifacts; the frontend use case is browser in the loop verification.

Sourceantigravity.google/docs/permissions, llms.txt (Browser, Browser Recordings) and use-cases/frontend; antigravity.google/docs/ide/browserread 2026-09-29

The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded

Product updates

·MCP / tool calling / APIPartially Verified

Any editor that supports the Agent Client Protocol can now use Antigravity CLI 1.5.0 as its agent, through a new server mode. Each session keeps its own model and thinking level and can be resumed without replaying its history. Auto mode also asks less often before safe commands that leave the terminal sandbox, such as installing registry packages or deleting files that version control can restore.

Bears on: Integrations

View source
·Deployment / data residencyPartially Verified

Remote agents are starting to roll out in Antigravity App v2.29.0 for Google AI Pro and Ultra subscribers. Conversations in a cloud project run in the cloud instead of on the user's computer, and they can also be opened on the web. The same release begins a gradual rollout of voice mode, built on Gemini Live, for handing the agent requests by voice.

Bears on: Deployment / data residency

View source
·Deployment / data residencyPartially Verified

Teams on Google Cloud Workstations inside a restricted VPC with no internet access can now start Antigravity for VS Code v1.11.0. In Cloud Shell and Cloud Workstations the extension uses the agy binary already installed there instead of downloading its own copy, which also speeds up startup.

Bears on: Deployment / data residency

View source
View all 12 Google Antigravity updates →Tracked since Jul 2026 · Verified from public vendor sources

Pricing

Free for individuals. Google AI Pro is $19.99/mo, and Ultra starts at $99.99/mo.

A monthly Google AI subscription per person, which sets the rate limits and includes a flexible AI credit pool on paid tiers. Organizations pay consumption based Google Cloud API pricing or use a Gemini Enterprise license.

Free tier

Included quota

The free Individual plan includes unlimited Tab completions and Command requests with basic weekly rate limits. Pro and Ultra add more generous rate limits and a flexible AI credit pool.

What is public

The free Individual plan gives access to Gemini and Claude models, with unlimited Tab completions and Command requests and basic weekly rate limits. Google AI Pro and Google AI Ultra add more generous limits and a flexible AI credit pool. An Organization plan runs through Google Cloud with consumption based API pricing and Gemini Enterprise. In the US, Google AI Pro is $19.99 a month and Ultra starts at $99.99, with a $199.99 option. The plans are listed at antigravity.google/pricing, and the subscription prices on Google's AI plans page.

Billing mechanics

A monthly Google AI subscription sets the rate limits and includes a flexible AI credit pool. Organizations pay consumption based Google Cloud API pricing or use a Gemini Enterprise license.

Cost watchouts

Frontier model use draws down rate limits and the AI credit pool faster. Organization use is billed on Google Cloud consumption.

Variable cost rationale

For individuals the price is a flat subscription, or nothing on the free plan. Use is held by rate limits rather than billed by the request, and frontier models draw those limits down faster. Pro and Ultra add a flexible AI credit pool for use beyond the rate limits, and no credit price is published. Organizations pay differently. Use through Google Cloud is billed on consumption based API pricing, so that bill grows with usage, or it runs under a Gemini Enterprise license.

Additional watchouts

Rate limits differ by tier, and neither the limits nor the price of AI credits is published as a figure.

Overage / add-ons

Pro and Ultra include a flexible AI credit pool beyond the rate limits. No credit price is published.

Sales call required

No, self serve available

Free / trial

Free Individual plan with unlimited Tab completions and Command requests, within basic weekly rate limits

Lowest paid plan

Google AI Pro at $19.99 per month

Commercial notes

Individuals subscribe through Google AI plans. Organizations buy through Google Cloud and Gemini Enterprise.

Key ambiguities

Per tier limits and credit values are not published.

Cancellation / refund

Google AI subscriptions are monthly. Organizations buy under the Google Cloud Terms of Service.

Support SLA / resale

Higher tiers add priority access and extended agent sessions. Organizations get support through Google Cloud and the Gemini Enterprise Agent Platform.

Missing data

Numeric rate limits per tier, the size of the AI credit pool and credit prices are not published.

Agentic Index verified 2026-10-25

Alternatives to Google Antigravity

The closest documented capability profiles to Google Antigravity among coding agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.

  • Capy12.5 / 14Matches Google Antigravity across all 14 documented capabilities
  • Cursor12.5 / 14Matches Google Antigravity across all 14 documented capabilitiesGoogle Antigravity vs Cursor →
  • Base4413.0 / 14Fuller documented coverage on Memory & State Persistence
  • Cline12.0 / 14A lighter documented profile than Google Antigravity
  • Factory13.0 / 14Fuller documented coverage on Knowledge Grounding & RAG
  • JetBrains AI12.0 / 14A lighter documented profile than Google Antigravity

Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded

Contact us

Found a vendor we missed? Have feedback on the index? We'd love to hear from you.