Google AntigravityvsKun
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
Google Antigravity
Google · AI-native IDE
#68MCP
Panel
6.03 spec wins
- Reliability
- 5.3
- Usefulness
- 6.7
- Cost
- 6.2
- Longevity
- 5.8
“Google's own IDE ships Claude and gpt-oss next to Gemini, the most honest model review Google has ever published.”
Kun
KunAgent · AI-native IDE
#84MCP
Panel
5.81 spec wins
- Reliability
- 5.7
- Usefulness
- 6.2
- Cost
- 6.5
- Longevity
- 4.8
“It writes your code and also your slide deck, which is the most accurate description of an engineering week ever shipped as a feature list.”
Spec by spec
| Spec | Google Antigravity | Kun |
|---|---|---|
| Architecture | ||
| Category | AI-native IDE | AI-native IDE |
| Runs | local | local |
| Platforms | macos, linux, windows | macos, linux, windows |
| Context window | not documented | not documented |
| Protocols | ||
| MCP client | Yes | Yes |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | Yes | Yes |
| Git operations | Yes | Yes |
| Browser control | YesA first-party browser subagent drives Chrome from a separate profile with allowlist/denylist controls (https://antigravity.google/docs/ide/browser/); IDE-side only, not in the agy CLI. | YesBrowser output is attached to tasks and screenshots are captured through browser automation. |
| Sandboxed execution | YesOS-level isolation rather than containers: "no virtual machines or Docker images to manage" - Linux uses kernel namespaces to isolate the filesystem, hide host processes and cut networking; macOS uses sandbox-exec with Seatbelt (SBPL) profiles. | No |
| Multi-agent orchestration | Yes | No |
| Headless / CI mode | Yes | No |
| Models | ||
| Backbone | Gemini 3.1 Pro, Gemini 3.8 Flash, Claude Sonnet 4.6, Claude Opus 4.6, gpt-oss-120b | OpenAI, Claude, Gemini, Ollama, DeepSeek, Kimi, GLM, Qwen, MiniMax |
| Bring your own model | YesSetting `modelProvider: "gemini"` in ~/.gemini/antigravity-cli/settings.json plus a GEMINI_API_KEY runs the CLI on your own key and billing instead of a signed-in Google account, and a model selector picks between Gemini, Claude and gpt-oss models in the IDE. | Yes |
| Local models | YesGOOGLE_GEMINI_BASE_URL sends model requests to any Gemini-compatible endpoint, so a self-hosted shim works as a mechanism; no Ollama or LM Studio integration is documented. | YesOllama ships as a named provider preset, and base URLs are editable so any OpenAI-compatible or Anthropic-compatible endpoint can be added. |
| Cost | ||
| Pricing model | mixed | byok |
| Starts at | $20/mo | n/a |
| Free tier | Yes | Yes |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | No | No |
| License | proprietary | PolyForm-Noncommercial-1.0.0 |
| GitHub stars | 2,446 | 6,324 |
Which one would each critic pick
| Critic | Google Antigravity | Kun | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 7.0 | 6.3 | Google Antigravity — Pick Antigravity if you want several agents running at once with a browser checking their work; pick Cursor if you want one agent that edits well and stays out of the way. |
| El Crítico | 6.3 | 5.8 | Google Antigravity — Agents drive your editor, terminal and a live Chrome with no container between them and your machine; the only isolation on offer is a git worktree. |
| El Profesor | 6.3 | 6.5 | Kun — The graphical client and the terminal client both attach to one local runtime, so threads, plans and approvals are shared state rather than two synchronised copies. |
| La Inversora | 7.3 | 5.0 | Google Antigravity — Google is bundling the IDE into Google AI Pro and Ultra subscriptions, distribution no startup on this board can match and a roadmap no buyer controls. |
| La Jefa | 5.3 | 4.8 | Google Antigravity — The Organization plan is consumption-based through Google Cloud with no seat price on the page, so the invoice for sixty developers is a forecast, not a number. |
| El Hacker | 4.0 | 6.5 | Kun — PolyForm Noncommercial, so I can read it and never ship a competing fork, but Ollama is a named preset and every base URL is editable. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.