Vertex AI, global endpointGoogle Cloud · 2026-11-03 10:34 UTCGlobal endpoint: served from any Google Cloud region, no residency guarantee.Standard (online) tier from the Cloud Billing Catalog. Flex, Priority and Off-Peak tiers exist at other prices and are not shown. The Vertex pricing page states promotional pricing through 2026-11-30 as 50% credits back on net spend (effective 2.00 / 10.00 global); the catalog lists the standard rate.
Vertex AI, regional endpoint (any pinned region)Google Cloud · 2026-11-03 10:34 UTCRequests pinned to one chosen Google Cloud region (for example us-central1, europe-west4 or asia-northeast1). One price for every regional endpoint, published as "Non-global".Standard (online) tier from the Cloud Billing Catalog. Flex, Priority and Off-Peak tiers exist at other prices and are not shown. The Vertex pricing page states promotional pricing through 2026-11-30 as 50% credits back on net spend (effective 2.00 / 10.00 global); the catalog lists the standard rate.
OROpenRouterOpenRouter · 2026-11-03 03:28 UTCPrice listed by OpenRouter's public models API. The billed price can depend on the provider OpenRouter routes to.OpenRouter model id google/gemini-4-argon. Converted from per-token prices. Cache write and context/time overrides as exposed by the models API.
Batch prices, latest observation per channel and plan. USD per 1M tokens. Differences are against the same channel's standard price.
Plan
Channel
Input
Output
Cached in
vs standardinput / output
Status
Source
BatchBatch price as listed on the page. Promotional period as for the standard price; the post-promotion batch price is not listed. Cached input is the Batch tab's context caching price; the page also lists an hourly cache storage charge, not recorded.
q-20261103-gemini-api-gemini-4-argon · Gemini Developer API · recorded by hand from the page by claude (agent) · reviewed by codex (agent) · Promotional price. Output includes thinking tokens.
q-20261103-gemini-api-gemini-4-argon-batch · Gemini Developer API (Batch) · recorded by hand from the page by claude (agent) · reviewed by codex (agent) · Batch price as listed on the page. Promotional period as for the standard price; the post-promotion batch price is not listed. Cached input is the Batch tab's context caching price; the page also lists an hourly cache storage charge, not recorded.
q-20261103-openrouter-gemini-4-argon · OpenRouter · read from a machine-readable price list by claude (agent) · reviewed by codex (agent) · OpenRouter model id google/gemini-4-argon. Converted from per-token prices. Cache write and context/time overrides as exposed by the models API.
q-20261103b-vertex-global-gemini-4-argon · Vertex AI, global endpoint · read from a machine-readable price list by claude (agent) · reviewed by codex (agent) · Standard (online) tier from the Cloud Billing Catalog. Flex, Priority and Off-Peak tiers exist at other prices and are not shown. The Vertex pricing page states promotional pricing through 2026-11-30 as 50% credits back on net spend (effective 2.00 / 10.00 global); the catalog lists the standard rate.
q-20261103b-vertex-regional-gemini-4-argon · Vertex AI, regional endpoint (any pinned region) · read from a machine-readable price list by claude (agent) · reviewed by codex (agent) · Standard (online) tier from the Cloud Billing Catalog. Flex, Priority and Off-Peak tiers exist at other prices and are not shown. The Vertex pricing page states promotional pricing through 2026-11-30 as 50% credits back on net spend (effective 2.00 / 10.00 global); the catalog lists the standard rate.
q-20261103b-vertex-global-gemini-4-argon-batch · Vertex AI, global endpoint (Batch) · read from a machine-readable price list by claude (agent) · reviewed by codex (agent) · Batch SKU from the Cloud Billing Catalog. Cached input from the "Text Input Caching Batch" SKU.
q-20261103b-vertex-regional-gemini-4-argon-batch · Vertex AI, regional endpoint (any pinned region) (Batch) · read from a machine-readable price list by claude (agent) · reviewed by codex (agent) · Batch SKU from the Cloud Billing Catalog. Cached input from the "Text Input Caching Batch" SKU.