Hyperscaler AI platform
How Google Vertex AI compares
Agent Platform generative pricing. Global endpoint rates from the first Gemini 4 Argon and Gemini 3.8 Flash tables.
Pricing page · Site · Open in compare
Later tables on the same page list different rates. Their tier labels did not survive extraction, so only the first global table for each model is stored.
Regions and compliance
Regions: not published / unverified
Compliance: not published / unverified
Token offerings
Gemini 4 Argon global (through 30 Nov 2026)
serverless_api · gemini-4-argon · open-weight not published / unverified
- Input / 1M
- 2cloud.google.comchecked 2026-10-20“Gemini 4 Argon* through November 30, 2026 | Input | Global | $2 | $0.1”
- Cached input / 1M
- 0.1cloud.google.comchecked 2026-10-20“Gemini 4 Argon* through November 30, 2026 | Input | Global | $2 | $0.1”
- Output / 1M
- 10cloud.google.comchecked 2026-10-20“Text output (response and reasoning) | Global | $10”
Context not published · tokens/s not published · TTFT not published
First global table on the page. The page also says non-global input is $2.2. A higher list rate starts 1 Dec 2026.
Gemini 3.8 Flash global (through 31 Dec 2026)
serverless_api · gemini-3.8-flash · open-weight not published / unverified
- Input / 1M
- 0.75cloud.google.comchecked 2026-10-20“Gemini 3.8 Flash* through December 31, 2026 | Input | Global | $0.75”
- Cached input / 1M
- 0.075cloud.google.comchecked 2026-10-20“Gemini 3.8 Flash* through December 31, 2026 | Input | Global | $0.75 | $0.075”
- Output / 1M
- 3.75cloud.google.comchecked 2026-10-20“Text output (response and reasoning) | Global | $3.75”
Context not published · tokens/s not published · TTFT not published
Compute SKUs
No GPU or node price is seeded.