Data platform

Snowflake Cortex AI

Model prices vs the direct API

snowflake.comChecked Sep 23, 2026

Pricing page Docs

Fee on tokens: 20%; Snowflake AI Credits per 1M tokens (input and output priced separately)
BYOK fee: No BYOK
Entry plan: Usage based
Hosting: Hosted

Dollar figures assume $2.00 per AI Credit, Global, On Demand, from Table 2(b) of the Snowflake Service Consumption Table (effective 2026-09-23); Regional AI Credits are $2.20

Models on Snowflake Cortex AI

7 model routes. 0 of 6 at the lab's own list price; the rest shown as a premium or discount.

ModelHere, in / outDirectPremium

Premium: blended price (3 input to 1 output) on this platform against the lab's own API. Regions are separate routes.

Features

Fallbacks
Load balancing
Caching
Guardrails
Observability
Rate limits
Budgets
Prompts
OpenAI-compatible
Bring your own key
Data residency
Private networking

What it costs on real workloads

Claude Sonnet 5$480.00vs $400.00 direct
GPT-6 Sol$480.00vs $400.00 direct
Gemini 3.8 Flash$180.00vs $150.00 direct
DeepSeek V4.1 Flash$32.40vs $27.00 direct

100M input and 20M output tokens a month, paying with gateway credits.

The terms, in their words

Snowflake's in-platform LLM functions and REST API that run Claude, GPT, Gemini, Llama, Mistral and other models on Snowflake-managed compute.

Markup: Snowflake AI Credits per 1M tokens (input and output priced separately)
Providers: Table 6(a) lists roughly 80 AI_COMPLETE models from Anthropic, OpenAI, Google, Meta, Mistral, xAI, Moonshot, Zhipu, Qwen, Snowflake and others

AI_COMPLETE - claude-sonnet-4-6 1.80 9.00 (AI Credits per one million Tokens, input / output)
Global $2.00 (On Demand AI Credit Pricing); Regional $2.20
Additionally, Cortex Inference may not be eligible for Capacity AI Credit Pricing.

Source: www.snowflake.com/legal-files/CreditConsumptionTable.pdf