Antigravity models: which plan can use Gemini 4 Argon?
Gemini 4 Argon is listed for Google AI Ultra and Enterprise, in the Antigravity app, CLI and IDE extensions; it is unavailable in the standalone Antigravity IDE. Gemini 3.8 Flash is listed across all four plan groups. Check the table for Claude 5.5 eligibility and choose a tested migration path before Claude 4.6 and GPT-OSS-120b leave on November 2.
Check your plan and migration path ↓1. Model availability by plan
This is the 2026-10-14 official Models matrix, with all ten reasoning models and three footnotes. Eligibility does not establish remaining quota or the choices currently visible in your account. Official matrix.
Argon has both a plan gate and a client gate. For an individual account, the listed plan is Google AI Ultra; Enterprise follows its own route. Use the Antigravity app, CLI or IDE extensions, not the standalone Antigravity IDE. Gemini app access under Google AI Pro does not grant Argon access in Antigravity.
Claude 5.5 requires non-trial Pro, Ultra or Enterprise. Free/Plus is excluded. Claude 4.6 is listed only for Free/Plus and Pro. The status column preserves the November 2 removal notice and Argon client restriction on mobile.
| Model | Free / Google AI Plus | Google AI Pro | Google AI Ultra | Enterprise | Status / removal |
|---|---|---|---|---|---|
| Gemini 4 Argon | Free / Google AI PlusNo | Google AI ProNo | Google AI UltraYes | EnterpriseYes | Status / removalApp / CLI / IDE extensions only; not standalone IDE |
| Gemini 3.8 Flash | Free / Google AI PlusYes | Google AI ProYes | Google AI UltraYes | EnterpriseYes | Status / removalListed |
| Gemini 3.7 Flash | Free / Google AI PlusNo | Google AI ProNo | Google AI UltraNo | EnterpriseYes | Status / removalListed |
| Gemini 3.6 Flash | Free / Google AI PlusNo | Google AI ProNo | Google AI UltraNo | EnterpriseYes | Status / removalListed |
| Gemini 3.1 Pro | Free / Google AI PlusNo | Google AI ProNo | Google AI UltraNo | EnterpriseYes | Status / removalListed |
| Claude Sonnet 5.5 (thinking) | Free / Google AI PlusNo | Google AI ProYes — non-trial only | Google AI UltraYes | EnterpriseYes | Status / removalListed |
| Claude Opus 5.5 (thinking) | Free / Google AI PlusNo | Google AI ProYes — non-trial only | Google AI UltraYes | EnterpriseYes | Status / removalListed |
| Claude Sonnet 4.6 (thinking) | Free / Google AI PlusYes | Google AI ProYes | Google AI UltraNo | EnterpriseNo | Status / removalRemoval: 2026-11-02 |
| Claude Opus 4.6 (thinking) | Free / Google AI PlusYes | Google AI ProYes | Google AI UltraNo | EnterpriseNo | Status / removalRemoval: 2026-11-02 |
| GPT-OSS-120b | Free / Google AI PlusYes | Google AI ProYes | Google AI UltraYes | EnterpriseNo | Status / removalRemoval: 2026-11-02 |
2. Choose migration candidates by your actual plan
Editorial selection guidance: begin with an eligible candidate for your current account and client, then validate your workflow. This is not a performance ranking or a promise of equivalent results. Dated availability source.
- Free / Google AI Plus: investigate Gemini 3.8 Flash for work that currently depends on Claude 4.6 or GPT-OSS. Neither Argon nor Claude 5.5 is listed.
- Google AI Pro: start with Gemini 3.8 Flash; non-trial subscriptions also list Sonnet 5.5 and Opus 5.5. A Pro trial does not qualify for those Claude models. Argon is not listed for Pro.
- Google AI Ultra: choose among Argon, Gemini 3.8 Flash and both Claude 5.5 models. Check the client before choosing Argon. GPT-OSS is scheduled for removal, and Claude 4.6 is already excluded.
- Enterprise: the table lists Argon, Gemini 3.8 / 3.7 / 3.6 Flash, Gemini 3.1 Pro and both Claude 5.5 models. Confirm your administrator’s enabled models and supported client; consumer subscription rules do not establish enterprise access.
There is no documented automatic replacement for a saved removed-model selection. Write down the workflow’s required behavior, tool permissions and acceptance checks before comparing candidates. Do not infer equivalent behavior from a model family or a newer version number. This checklist does not require a plan purchase.
3. A new dropdown selection does not hot-swap a running turn
Google says the reasoning-model choice persists between messages in a conversation. When you change it during execution, the current turn continues using the previous selection until its steps finish or you cancel that execution. The change is not a mid-turn hot swap. Turn behavior.
Editorial operating rule: let the current turn settle, or deliberately cancel it and inspect work already done. Confirm the new selection before starting the next task. Do not launch duplicate work merely because a dropdown changed.
Post-removal saved-conversation behavior: UNKNOWN. The reviewed official pages do not specify how a saved selection of a removed model will be mapped, rejected or replaced after November 2. Do not assume automatic migration to Claude 5.5 or Gemini. Inventory important saved conversations now and inspect the active choice when reopening them.
For Implementation Plans, Review/Proceed and plan policy, use the separate Antigravity /plan guide. Those controls answer a different question from model eligibility.
4. Separate model access, quota and paid overages
The Plans page gives Pro a five-hour refresh until its weekly limit is reached, and Ultra a five-hour refresh plus weekly limits. Other consumer plans refresh weekly. Allowances depend on capacity and work performed and can change; this is not a fixed prompt-count promise.
The Models page illustrates separate “Gemini Models” and “Claude and GPT models” usage groups. That display does not establish every cross-model pooling rule or prove your account has the sample balance. Exact pool allocation and your remaining allowance are UNKNOWN until checked for the account. Do not infer that Claude 5.5 always consumes quota faster than Gemini, or that switching models resets a limit.
For Pro/Ultra, the Plans page describes credits above baseline and AI Credit Overages: Never / Always. Never waits for baseline refresh; Always permits credit use after exhaustion. A listed model can still be quota-limited. Our editorial recommendation is to inspect the current setting before testing, not to enable spending automatically.
The consumer Plans page’s no-BYOK statement concerns extra rate limits. Enterprise has a separately documented connection and billing path; do not merge the two contracts.
Keep the product and billing paths distinct. A Gemini app plan label is not proof of Antigravity entitlement. Direct Gemini API Argon access has its own API usage and billing conditions; a consumer subscription is not an API credit balance. See the current API access guidance. Before any test, record which product, identity and billing route will actually handle it.
5. Refresh the account view before selecting a destination
In the CLI, Google documents /usage (alias /quota) as refreshing model configuration and quota status from the backend, then opening the quota panel. The usage guide’s example model names are illustrative; the current Models matrix above owns the dated availability list. CLI usage reference.
AgentSkillsHub editorial checks:
- Confirm the signed-in account and whether its plan is Free/Plus, Pro trial, non-trial Pro, Ultra or Enterprise.
- Confirm the client: Antigravity app, CLI, IDE extension or standalone IDE. Argon is excluded from the standalone IDE. Open the current model selector and record the choices actually offered. A model name in an old message is not proof of present access.
- Refresh usage/configuration where your client supports it; record remaining quota and relevant refresh timing.
- If the selector differs from the dated table, check the official notice and account/client route before diagnosing it as a product bug. Google notes that app releases roll out gradually; that is not proof of why one particular model is missing.
- Preserve the observed discrepancy and ask the appropriate administrator or support channel when eligibility remains unclear. Do not invent a fallback or bypass an account restriction.
No account configuration was changed while preparing this guide.
6. Keep Enterprise choices and billing separate
Google documents Enterprise access through either a Cloud project/API with consumption billing or a Gemini Enterprise Standard/Plus license with included quotas and managed overages. The supported route includes the Antigravity app, CLI and IDE extensions; the standalone Antigravity IDE is not currently supported for enterprise deployment. Enterprise guide.
Editorial migration check: ask the administrator which account, license/project, client and overage policy your workflow uses. Record the Gemini and Claude 5.5 choices actually enabled on that route. The Models matrix now includes both Claude 5.5 models for Enterprise; this is not a promise that an administrator has enabled them. Do not apply consumer Pro non-trial access or its quota schedule to an Enterprise user by analogy.
No Enterprise account or billing configuration was tested here.
7. Before November 2: verify one representative workflow
AgentSkillsHub editorial migration checklist — not a Google mandatory procedure. Use the dated removal notice as the trigger and your own acceptance criteria as the decision.
- Inventory dependencies: find conversations, saved instructions and workflows naming Sonnet 4.6, Opus 4.6 or GPT-OSS-120b. Record the task owner and last known selection.
- Confirm the account plan: include Pro trial status, Enterprise route and the client restriction for Argon. Do not infer it from a teammate’s account.
- Open the current selector: record eligible destination candidates and any discrepancy against the official table.
- Choose a test candidate: write down why it is eligible and what result the workflow must preserve; avoid declaring a universal “best model.”
- Run a representative task: use a safe branch or synthetic input and an approved quota/spending boundary. This is a proposed test, not one we ran.
- Compare actual results: behavior, tool use, code diff, latency and quota impact where observable. Missing measurements remain unknown. Review the diff and output rather than relying on a success message.
- Update team instructions: replace removed-model names only after acceptance. Record the tested plan/model, unresolved differences and a fallback available to that same plan.
- Prepare before the deadline: do not leave high-value workflows until November 2. Recheck Google’s notice and the account selector shortly before the change.
If no candidate passes, keep the failure evidence and arrange a manual path or a separate decision. A removal notice does not authorize a subscription change, broader permissions or automatic production actions.
8. Example: a Free-plan workflow still names Sonnet 4.6
EDITORIAL EXAMPLE — NOT RUN. A developer reopens a saved conversation whose runbook says to use Sonnet 4.6 for a small bug fix. Their current account is Free, so “switch to Claude 5.5” is not an eligible default response.
They record the branch revision and the bug’s acceptance check, inspect today’s selector, and consider Gemini 3.8 Flash, the Gemini option listed for Free in this review for a synthetic reproduction. They compare the proposed code diff, tests, tool choices and observable quota use. A different answer or latency is evidence to review, not proof that the replacement is better or worse for every task.
Only after the task passes do they update that runbook’s model instruction. They retain an explicit fallback plan and still inspect the selection when reopening old conversations after the removal. Automatic post-removal mapping remains unknown.
This example changed no account, called no model and consumed no quota.
9. Additional models are not selectable reasoning models
The official Additional models section names Nano Banana 2 for the generative-image tool, including mockups, application images and diagrams. Google describes these additional stack models as not customizable. Keep that auxiliary role separate from the reasoning-model selector and plan matrix.
This guide does not infer more image-model choices, interchangeable quotas or a model-quality ranking. The next useful action is to check your actual plan and selector, then work through the migration checklist.
FAQ
Can Google AI Pro use Gemini 4 Argon in Antigravity?
No. Argon is listed for Google AI Ultra and Enterprise. Gemini app availability under Pro does not establish Antigravity access, and API usage follows a separate billing route.
Why is Argon missing from the standalone Antigravity IDE?
Google explicitly excludes the standalone Antigravity IDE. Argon is listed for the Antigravity app, CLI and IDE extensions, subject to Ultra or Enterprise eligibility and account availability.
Can a Google AI Pro trial use Claude 5.5 in Antigravity?
No. The Pro footnote excludes trial subscriptions. The table also lists Claude 5.5 for Ultra and Enterprise, while Free/Plus is excluded. Check the actual account and administrator configuration.
Which models are scheduled to leave on November 2, 2026?
Google lists Claude Sonnet 4.6 (thinking), Claude Opus 4.6 (thinking) and GPT-OSS-120b. The reviewed notice does not give an exact removal hour or timezone.
Can a Free or Plus user simply replace Claude 4.6 with 5.5?
No. Free/Plus does not list Claude 5.5. Gemini 3.8 Flash is the Gemini candidate listed for this group; validate your workflow before changing its runbook.
Does Ultra include Claude 4.6?
No. The 2026-10-14 table excludes both Claude 4.6 models from Ultra. It lists Argon, Gemini 3.8 Flash, both Claude 5.5 models and GPT-OSS, with GPT-OSS scheduled for removal.
Does changing the model switch an agent in the middle of its turn?
No mid-turn hot swap is documented. The current turn uses the previous selection until its steps finish or you cancel execution. The selected model persists between messages in the conversation.
What happens to a removed model saved in an old conversation?
UNKNOWN. The reviewed official pages do not specify the exact post-removal fallback. Do not assume automatic migration to Claude 5.5 or Gemini; inspect the selection before continuing important work.
Are Gemini and Claude/GPT quotas interchangeable?
The model picker shows separate usage groups, but the reviewed documentation does not establish every shared-pool rule. Check current account usage; do not infer a reset or a fixed consumption ratio from switching models.
Did AgentSkillsHub test every plan or run a migration benchmark?
No. This is a source-checked guide. We changed no plan, did not switch through every model and consumed no model quota. The checklist and example are editorial material.
Primary sources
Source-checked 2026-10-14. Documentation review is not an account or migration benchmark. Examples and checklists here are editorial material.
- Google: reasoning model matrix, footnotes and turn behavior
- Google: plans, rate limits and AI credit overages
- Google: CLI model quotas and configuration refresh
- Google: Enterprise connection and supported surfaces
- Google: Antigravity changelog and rollout notes
- Google: Gemini 4 Argon API model reference
- Google: latest Gemini API model and access guidance