- BFSIBanking, Financial Services & Insurance
- HealthcareComing soon
- ManufacturingComing soon
- Retail & E-CommerceComing soon
Snowflake AI Data Cloud
Your data sits in five systems and every team keeps its own copy. Analytics, apps and AI should read the same tables — Snowflake AI Data Cloud keeps data, warehouses, pipelines, apps and AI agents in one managed account, billed in credits, with AWS Mumbai and Azure Pune priced the same as US East.
Buy through TechBag
Same software. Better outcome — at a lower cost.
How it’s rated
Full scoreboard ↓Quick answer
The Snowflake platform family
This page covers Snowflake AI Data Cloud — the platform and its four editions, including Cortex AI, Snowpark and Openflow. The rest:
Quick facts
30-second orientation- Product
- One managed platform for warehousing, data engineering, apps and AI agents, billed in credits
- Maker
- Snowflake Inc. (NYSE: SNOW); CEO Sridhar Ramaswamy since February 2024
- Editions
- Standard, Enterprise, Business Critical and Virtual Private Snowflake (VPS)
- Price
- $2 / $3 / $4 / $6 per credit by edition in Mumbai and Pune, the same as US East
- Compute
- XS warehouse = 1 credit an hour, doubling per size to 512 at 6XL; per second after 60 s
- Storage
- $23 per TB a month on demand in Indian regions; capacity tiers fall to $13.80
- AI
- Cortex AI uses AI credits ($2.20 regional, $2.00 global), priced per million tokens
- Scale
- Product revenue $1.49B in Q2 FY2027, up 37%; 828 customers above $1M a year
- India
- AWS Mumbai (ap-south-1) and Azure Central India (Pune); no Google Cloud region in India
- In India via
- TechBag — region and edition choice, warehouse sizing, quote in INR with GST
Understand cloud data platforms before you buy one
Most product pages skip this. We start here — so you buy a capability, not a buzzword.
What is a cloud data platform?
Storage and compute are bought separately, so many teams can query the same data without making copies.
A fixed cluster sized for peak vs Snowflake AI Data Cloud — the honest table
What consolidation actually replaces, dimension by dimension.
| Dimension | A fixed cluster sized for peak | Snowflake AI Data Cloud |
|---|---|---|
| Sizing compute | A cluster bought for the busiest day of the year | Warehouses from XS to 6XL, billed per second |
| Teams competing | One shared cluster and one long queue | A warehouse per team over the same tables |
| Adding AI | Data exported to a separate ML stack | Cortex AI and Snowpark inside the account |
| Where data is stored | Wherever the appliance was racked | AWS Mumbai or Azure Pune, chosen per account |
| How you pay | Licences and hardware up front | Credits on demand each month, or prepaid capacity |
| What it is NOT | — | Self-managed software, a flat monthly fee, or on Google Cloud in India |
The cheapest test is the free trial: load one subject area into a Mumbai account, replay a month of reports, and read the credits used.
The five pieces of the platform
Vendors love diagrams; buyers need to know what they’re actually operating. Here’s the whole platform, demystified.
Storage
Loaded data is kept in the cloud region the account was created in and billed per terabyte-month: $23 on demand in Mumbai or Pune, falling to $13.80 on capacity tiers.
Warehouses
Compute runs in warehouses you size from XS at 1 credit an hour to 6XL at 512, billed per second after the first minute; a Gen2 XS costs 1.35 credits an hour on AWS.
Horizon
Horizon governance comes with every account at no extra line item; regulated buyers step up to Business Critical for Tri-Secret Secure and its disaster-recovery tooling.
Cortex and apps
Cortex AI functions, Snowpark code, Streamlit apps, notebooks and Snowflake ML run inside the account; Cortex draws AI credits per million tokens, the rest platform credits.
Storage per terabyte-month, warehouses per second — with Cortex AI, Snowpark and Openflow running in the same account.
Nine capabilities. Store, run, build.
Snowflake AI Data Cloud is one managed platform where storage, compute and AI each run on their own meter.
Four editions, one platform
Standard is the entry point; Enterprise adds multi-cluster compute and longer Time Travel; Business Critical adds Tri-Secret Secure.
Ingestion inside the account
Openflow, built on Apache NiFi, brings data in from other systems; run in your own cloud it costs 0.0225 credits per vCPU-hour.
Transactions beside analytics
Unistore hybrid tables keep operational rows in the same account as analytic tables, paid for from the same platform credits.
Compute sized in steps
Warehouses run from XS at 1 credit an hour to 6XL at 512, each size doubling the last, billed per second after a 60-second minimum.
A newer warehouse generation
Gen2 warehouses cost 1.35 credits an hour at XS on AWS and 1.25 on Azure; time your own queries before paying the higher rate.
Busy warehouses scale out
From Enterprise up, a warehouse can run as several clusters at once, so heavy report hours do not need one oversized warehouse.
AI functions called from SQL
AISQL functions, Cortex Search, Cortex Analyst and Cortex Agents run in the account, metered in AI credits per million tokens.
Your code next to the data
Snowpark, notebooks and Snowflake ML let engineers run their own code and model training inside the account on platform credits.
Datasets and apps to share
Marketplace and Native Apps let an account publish or acquire datasets and applications, with Streamlit for building simple front ends.
Watch Snowflake AI Data Cloud in action
A September 2026 customer story from Zeta Global, Snowflake’s mid-2025 platform update and Summit 2025 keynote, and a 2021 introduction from the Data Cloud era.
How Zeta Global Powers AI Agents With Snowflake AI Data Cloud
Zeta Global on running AI agents on the platform, published on Snowflake’s own channel.
Snowflake AI Data Cloud Updates
A mid-2025 round-up of platform changes, from before the June 2026 product renames.
Snowflake Summit 2025 Platform Keynote
The 2025 platform keynote; Snowflake Intelligence has since been renamed Snowflake CoWork, and Cortex Code is now Snowflake CoCo.
Introduction to the Snowflake Data Cloud
A 2021 introduction from when the platform was called the Data Cloud; check prices and features against today’s pages.
Want a live, India-context walkthrough for your environment?
Book a guided demo →Every team wants its own compute, and nobody wants another copy of the data. Snowflake gives each team a warehouse over the same tables.
Here’s what genuinely sets it apart — and exactly where it stops.
Indian regions at US East prices
A credit costs the same in AWS Mumbai and Azure Central India (Pune) as in US East: $2 on Standard up to $6 on VPS. Singapore charges $2.50 to $7.50 for the same editions, so keeping the account in India avoids an Asia-Pacific premium.
Compute billed only while it runs
Each warehouse size has a fixed rate, from 1 credit an hour at XS to 512 at 6XL, charged by the second after a 60-second minimum. Storage is metered apart, so every team can run its own warehouse over the same data.
Warehouse, pipelines and AI in one account
Openflow, Unistore hybrid tables, Snowpark, Streamlit, Marketplace and Cortex AI share one account under Horizon governance, so an app or AI function needs no copy in another product. Snowflake took over 330 capabilities to GA in the first half of fiscal 2027.
Where it stops
Credits make starting cheap and budgeting hard: an oversized warehouse left running turns straight into spend, and Cortex AI has its own AI-credit meter. There is no self-managed edition, no Google Cloud region in India, and password-only service users break as MFA enforcement ends in October 2026.
The numbers behind the platform
What your Snowflake AI Data Cloud rollout looks like
Pick the region and edition
Choose AWS Mumbai or Azure Pune, and decide whether you need Enterprise multi-cluster or Business Critical’s Tri-Secret Secure.
Open a trial with guardrails
Start a trial account, create one XS or S warehouse per team, and set resource monitors before any real data arrives.
Load and replay real work
Bring in one subject area through Openflow or bulk loads, replay last month’s reports, and log the credits each one used.
Right-size, then add AI
Resize warehouses from the usage data, then pilot one Cortex AI function and track its AI credits on their own line.
Choose on demand or capacity
Turn two months of usage into a forecast, move service users off passwords, and pick on-demand or prepaid capacity.
The review scoreboard
Modelled on Gartner Peer Insights structure. *Counts and breakdowns are illustrative pending verified review collection.
Quick poll — what’s driving your evaluation?
Talk to an advisor“Month-end close used to queue behind marketing’s jobs. A separate warehouse per team on the same tables ended that fight.”
“Keeping the account in Mumbai at US East credit rates made the residency talk with our risk team very short.”
“A forgotten Large warehouse ate a week of budget in two days. Put resource monitors in place before handing out admin rights.”
“Our first Cortex summarisation job was cheap to build and costly to run over every ticket. Watch the AI-credit line separately.”
“Leaving our on-premises appliance, loading data was easy; rewriting the stored procedures took most of the project.”
“Business Critical was worth the step up for us. Tri-Secret Secure was the first control our auditors asked to see.”
Where everyone sits — the grids
Analyst firms bury this view behind paywalls, and G2 retired its Grid. So here’s TechBag’s synthesis of the cloud data platform market — tap any vendor to see why it sits where it does.
TechBag Cloud Data Platform Grid
Execution strength vs product vision — the classic market map, minus the paywall.
Printed per-credit prices; Mumbai and Pune at US East rates.
Price Transparency × Workload Breadth
The grid nobody publishes — how plainly the unit price is printed vs how many kinds of work run on one platform.
Credit rates printed per edition; warehouse, pipelines, apps and AI.
Positions are TechBag’s illustrative synthesis of public review-platform data and vendor documentation — not a reproduction of any analyst graphic. Verify before relying on it.
Snowflake AI Data Cloud vs the cloud data platform field
Set beside Databricks, Google BigQuery, Microsoft Fabric, Amazon Redshift and SAP Business Data Cloud — on pricing units, entry price, sizing, AI, India storage and exit.
| Dimension | Snowflake AI Data Cloud | Databricks | Google BigQuery | Microsoft Fabric | Amazon Redshift | SAP Business Data Cloud |
|---|---|---|---|---|---|---|
| What it is | Cloud data platform | Lakehouse platform | Serverless warehouse | SaaS suite on OneLake | AWS’s data warehouse | SAP’s data layer |
| Deployment | AWS, Azure or GCP | AWS, Azure or GCP | Inside Google Cloud | Azure capacity | AWS only | SaaS run by SAP |
| Pricing model | Credits by edition | DBUs, pay as you go | Bytes or editions | Capacity-unit hours | RPU-hours or nodes | Quoted per customer |
| Published entry price | $2 per credit | No DBU rate shown | $6.25 per TiB | F2 about $262/month | $0.375 per RPU-hour | Not published |
| Compute sizing | XS to 6XL warehouses | One unit for all | No clusters to size | F-SKU named by CUs | Base 4–1,024 RPUs | No published unit |
| Billing granularity | Per second, 60 s min | Per second | Per query scanned | Per CU-hour | Per-second RPU use | Subscription term |
| Free trial or tier | Trial with credits | 14 days + Free Edition | 1 TiB a month free | No free tier listed | $300 for 90 days | None documented |
| Included vs add-on | Storage, AI apart | Pay per product used | Storage billed apart | OneLake storage extra | Storage, Spectrum extra | Datasphere, Databricks |
| AI and agents | Cortex AI in the account | Built for data and AI | Vertex AI, Gemini | Copilot built in | SQL ML via SageMaker | Grounds SAP Joule |
| Integrations and sharing | Openflow, Marketplace | Read directly by others | Queried where it sits | Power BI on OneLake | Zero-ETL, no fee | SAP apps plus others |
| India storage region | Mumbai and Pune | AWS Mumbai | Mumbai and Delhi | Central India $0.20/CU | Mumbai and Hyderabad | Not published |
| Buying and billing | On demand or prepaid | Card or commitment | On the GCP bill | Azure subscription | On the AWS bill | SAP contract |
| Lock-in and exit | Managed-only service | Contracts span clouds | Google Cloud only | Microsoft estate pull | AWS-bound | SAP-shaped |
| Best fit | Multi-source analytics | Engineering-heavy teams | Analytics on GCP | Microsoft-first shops | AWS-centred estates | SAP-centred businesses |
Which approach fits you?
Honest fit signals — because the fastest way to lose your trust is to pretend one product wins every scenario.
Choose Snowflake AI Data Cloud if…
- ✓You want one managed platform for the warehouse, ingestion, apps and AI, with nothing to install, size up front or patch
- ✓Data has to be stored in India and you would rather pay US East credit rates than an Asia-Pacific premium
- ✓Several teams need their own compute over shared tables, each warehouse sized separately and billed by the second
Compare alternatives if…
- ✓Your engineers live in notebooks and Spark and want a lakehouse first — Databricks is Snowflake’s main rival
- ✓You are committed to one cloud’s own stack — BigQuery on Google Cloud, Fabric on Microsoft, Redshift on AWS
- ✓Most of your data is SAP’s and its business meaning matters most — look at SAP Business Data Cloud
Do not expect…
- ✓A self-managed or on-premises edition, or a Google Cloud region in India
- ✓A fixed monthly bill: credits, storage, egress and AI credits each move on their own
- ✓Password-only service logins to keep working once MFA enforcement completes in October 2026
Snowflake AI Data Cloud is one of 25 databases & data tools products TechBag carries. The Databases & Data Tools guide narrows them to a shortlist and shows the reasoning. →
What does running your own data clusters cost you?
Drag the sliders (data pipelines in production; data-engineer hour cost). Estimates model engineering time spent sizing clusters, planning capacity and keeping servers patched, at an assumed 1.5 hours per pipeline a year, with 70% of it removed by a managed platform billed per second. Both figures are assumptions. Illustrative.
Loaded cost = salary + overheads per productive hour. Illustrative only — your TechBag quote models your actual environment and modules.
Three ways to consume it
Published: Snowflake prints per-credit rates by edition, and AWS Mumbai and Azure Central India (Pune) match US East — $2 Standard, $3 Enterprise, $4 Business Critical, $6 VPS — plus $23 per TB a month for on-demand storage. An XS warehouse uses 1 credit an hour; Cortex AI is billed in AI credits. Pay on demand monthly or prepay capacity. TechBag forecasts credits from a trial, then quotes in INR with GST.
On Demand
Best for pilots and uneven workloads
- Standard $2, Enterprise $3 per credit in India
- Billed monthly in arrears, per second of use
- Storage $23 per TB a month
+ Platform add-ons
Best for a broader rollout
- Scoped to your estate
- Add-on modules as needed
- Phased, right-sized deployment
Capacity
Best for steady, forecastable spend
- Credits prepaid under a contract
- Storage tiers down to $13.80 per TB
- Business Critical $4, VPS $6 per credit
Buy it for less — TechBag pricing beats list
Whatever the list prices above, TechBag negotiates a significantly better deal — with GST-compliant INR invoicing and local support. Ask us for your discounted quote.
Get an India-ready quote
Tell us your requirements and current tools — we’ll model it against what you spend today.
The 8 questions to ask every vendor
Take this into your next vendor call — including ours.
Will the account sit in AWS Mumbai or Azure Central India (Pune)? Google Cloud has no Snowflake region in India.
Do you need multi-cluster compute (Enterprise), or Tri-Secret Secure and DR features (Business Critical)?
Which size does each workload need? Credits an hour double with every step, from 1 at XS to 512 at 6XL.
Who sets resource monitors and suspend rules before teams are allowed to create their own warehouses?
Will Cortex AI be used, and who watches the AI-credit meter that runs apart from platform credits?
Will data leave the region? Transfer out of AWS Mumbai costs $60 per TB to another AWS region and $90 beyond.
Have password-only service users moved to key-pair, OAuth or tokens before MFA enforcement ends in October 2026?
On demand or prepaid capacity? Ask for the INR figure with GST next to the USD credit rate for each edition.
Questions buyers ask
Ready to evaluate Snowflake AI Data Cloud?
Count the pipelines your team keeps running first, or let a TechBag advisor size warehouses from a trial and quote the credits in INR.
Stats, ratings, review counts and pricing are illustrative and sourced from public materials; verify before purchase.