Gemini FamilyGeneral TierTextGoogle Cloud Vertex AI
Gemini 4 Argon Cyber Permissive
Gemini 4 Argon Cyber Permissive pricing on Google Cloud Vertex AI.Rates are normalized to USD per 1 Million Tokens ($/1M). Vertex AI publishes one global rate per price type: there is no regional price list to browse. Read more about Google.
Vertex AI rates (global)
| Rate | USD per 1M tokens |
|---|---|
| Input | $4 |
| Output | $20 |
| Cache read | $0.2 |
| Batch input | $2 |
| Batch output | $10 |
| Input, long context | — |
| Output, long context | — |
| Output, thinking | — |
Headline rates are the standard tier; the file's batch tier is listed separately. Long-context and thinking rates apply where the model publishes them. A dash means Google publishes no rate for that price type, not a zero.
What to know
- Vertex AI rates are global: unlike Amazon Bedrock, which prices each model per region, Google publishes one rate that covers every region where the model runs.
- Claude is not shown on Vertex: Anthropic's models are priced there per geography and context band, and this catalog publishes them on Amazon Bedrock only.
- Compare Bedrock models on the model comparison and the token calculator.
Esc
Search instance types (e.g. "m5"), volumes, databases, AI models and comparisons.