$0bandwidth
$0/GB | $0/seat | Owned infrastructure
Amazon Web Services
$0.087/GB effective egress + per seat$345/month
Google Cloud
$0.093/GB effective egress + per seat$340/month
Microsoft Azure
$0.084/GB effective egress + per seat$344/month
All cloud providers charge you more when your app succeeds. We don't! With Catalyst, ingress and egress are $0, whether it's 1 TB or 500 TB.
We own our infrastructure, so data movement is our cost to absorb, not yours to pay. That means you can build for scale and move data freely between services, regions, and users without ever paying for the distance.
Calculate your bandwidth savingsWhy pay every time your data moves?With Catalyst, you don't.
In a typical user request, data crosses at least three vendor boundaries: app, database, and storage, and each one charges egress on the way out.
Data Transfer Cost Comparison
500 TB Data Transfer / MonthMove 500 TB every month. Pay $0 for data transfer with Catalyst.
| Platform | Monthly Cost | Yearly Cost |
|---|---|---|
| AWS S3 | $29,482.20 | $353,786 |
| GCP Cloud Storage | $41,277.32 | $495,328 |
| Microsoft Azure | $29,369.86 | $352,438 |
| Catalyst | $0 | $0 |
| Tier | Free tier | Tier 1 | Tier 2 | Tier 3 | Tier 4 |
|---|---|---|---|---|---|
| Volume range | Initial free usage | Up to ~10 TB | Mid-scale usage | Higher-scale usage | Very high-scale usage |
| AWS S3 (US East/us-east-1) | First 100 GB/month | 100 GB–10 TB → $0.090/GB | 10 TB–50 TB → $0.085/GB | 50 TB–150 TB → $0.070/GB | Over 150 TB → $0.050/GB |
| GCP Cloud Storage (us-central1/us-east1). GCP prices in GiB (gibibytes); 1 GiB ≈ 1.074 GB. | First 1 GiB/month | 1 GiB – 1 TiB → $0.120/GiB | 1 TiB – 10 TiB → $0.110/GiB | Over 10 TiB → $0.080/GiB | - |
| Microsoft Azure (Zone 1, Premium Global Network) | First 100 GB/month | 100 GB–10 TB → $0.087/GB | 10 TB–50 TB → $0.083/GB | 50 TB–150 TB → $0.070/GB | Over 150 TB → $0.050/GB |
| Catalyst | $0 | $0 | $0 | $0 | $0 |
The double billing problem
100 GB of user traffic / month
Typical Stack
Bytes arrive at the app host. Inbound traffic is free — no meter ticks yet.
Catalyst
Same story. Ingress is free on every cloud, including Catalyst.
Typical Stack:$0
Catalyst:$0
Typical stack
3 vendor boundaries → 3 egress bills
3 billed hops
User
Ingress is free everywhere$0.00
App host
Leaves Vercel network → user+ $12.00
Database
Leaves DB network → app+ $9.00
Storage
Leaves S3 network → app+ $12.00
Running cost
Vercel + managed DB + S3 · final:$33.00
$0.00
Catalyst
Same path, one network, zero billed hops
0 billed hops
User
Ingress is free everywhere$0.00
App host
Same platform, no boundary crossed.$0.00
Database
Same platform, no boundary crossed.$0.00
Storage
Same platform, no boundary crossed.$0.00
Running cost
One platform$0 egress
$0.00
See how this plays out in real stacks
A 3-vendor SaaS stack
The problem
A SaaS app built across three providers pays egress every time the stack talks to itself. The app queries Supabase, retrieves files from S3, and delivers responses via Azure. Each hop crosses a provider boundary. Pulling 100MB from the database and sending that same 100MB to the browser is billed twice, even if the data never left the app.
Before Catalyst
Daily egress x 365
Supabase is its own billing perimeter. Any byte leaving this dashed border is metered as egress.- AWS S3 is its own billing perimeter. Any byte leaving this dashed border is metered as egress.
Azure is its own billing perimeter. Any byte leaving this dashed border is metered as egress.
After Catalyst
$0 egress
- Stratus
- Slate
- NoSQL
Media platform startup
The problem
A video platform built across three providers pays egress at every stage of the pipeline. Uploads pulled from S3, transcoded on GCP, then thumbnails pushed to Cloudflare. Each step crosses a provider boundary and starts a new meter. Every video processed means egress from S3 and egress from GCP before a single user has watched anything. At 50,000 uploads/month, per-video cents become thousands of dollars in transfer fees alone.
Before Catalyst
Daily egress x 365
- AWS S3 is its own billing perimeter. Any byte leaving this dashed border is metered as egress.
Google Cloud is its own billing perimeter. Any byte leaving this dashed border is metered as egress.
Cloudflare CDN is its own billing perimeter. Any byte leaving this dashed border is metered as egress.
After Catalyst
$0 egress
- Stratus
- Slate
- DataStore
The ETL pipeline
The problem
A nightly ETL job moves gigabytes from S3 to Snowflake to a dashboard server. Each sync crosses egress boundaries, starting a meter. The data belongs to the team running the pipeline. They just keep paying to move it around their own stack. By year end, the egress bill exceeded what the actual compute cost to run the analysis. They just kept paying to look at it.
Before Catalyst
Daily egress x 365
- S3 Data Lake is its own billing perimeter. Any byte leaving this dashed border is metered as egress.
Snowflake is its own billing perimeter. Any byte leaving this dashed border is metered as egress.
DigitalOcean is its own billing perimeter. Any byte leaving this dashed border is metered as egress.
After Catalyst
$0 egress
- Stratus
- Slate
- Serverless functions
eCommerce backend
The problem
A single customer page load crosses three provider boundaries. Auth check to Firebase, product fetch from Atlas, inventory sync from Heroku. Three round trips, three egress meters for one page loaded. The app isn't doing anything unusual. It's just built the way most apps are built. Multiply that by millions of page loads per month, and the egress bill grows significantly.
Before Catalyst
Daily egress x 365
Firebase is its own billing perimeter. Any byte leaving this dashed border is metered as egress.
MongoDB Atlas is its own billing perimeter. Any byte leaving this dashed border is metered as egress.
Heroku is its own billing perimeter. Any byte leaving this dashed border is metered as egress.
After Catalyst
$0 egress
- Slate
- NoSql
- Stratus
Why Catalystcan offer $0 egress?
We own the infrastructure
Catalyst runs on hardware we own and operate, not capacity rented from AWS, GCP, or Azure. Data movement is an internal operational cost we absorb. That's structurally different from a platform that resells hyperscaler capacity and passes transfer fees to you.
Single infrastructure model
Your app, database, storage, and auth live on the same infrastructure. There are no service boundaries to cross and no egress events to trigger. Data moves internally, the way it should.
True pay-per-use
You pay for compute and storage, not for data moving between them, for serving users, or for your stack talking to itself.
Egress fees aren't infrastructurecost. They're a lock-in mechanism.
When bandwidth costs money, architecture decisions get made for the wrong reasons. The architecture you'd build if data were free is usually faster, more resilient, and easier to debug. Catalyst lets you build that one.
Predictable by design. No surprise line items.
Bandwidth charges are one of the hardest things to forecast on a cloud bill. Catalyst replaces that uncertainty with a guarantee: $0 for data movement, regardless of volume.
- Hard spend caps: When you hit your configured budget limit, Catalyst throttles compute and platform services instead of billing overages.
- No per-seat fees that creep as the team grows.
- No bandwidth tiers. Pay $0 whether you move 1 TB or 500 TB.
Cumulative Egress CostOver Time
Based on A 500 TB egress calculation, assuming monthly egress remains constant. Costs accumulate annually for hyperscalers, while Catalyst Stratus has zero egress.
| Tier | Free tier | Tier 1 | Tier 2 | Tier 3 | Tier 4 |
|---|---|---|---|---|---|
| Volume range | Initial free usage | Up to ~10 TB | Mid-scale usage | Higher-scale usage | Very high-scale usage |
| AWS S3 (US East/us-east-1) | First 100 GB/month | 100 GB–10 TB → $0.090/GB | 10 TB–50 TB → $0.085/GB | 50 TB–150 TB → $0.070/GB | Over 150 TB → $0.050/GB |
| GCP Cloud Storage (us-central1/us-east1). GCP prices in GiB (gibibytes); 1 GiB ≈ 1.074 GB. | First 1 GiB/month | 1 GiB – 1 TiB → $0.120/GiB | 1 TiB – 10 TiB → $0.110/GiB | Over 10 TiB → $0.080/GiB | - |
| Microsoft Azure (Zone 1, Premium Global Network) | First 100 GB/month | 100 GB–10 TB → $0.087/GB | 10 TB–50 TB → $0.083/GB | 50 TB–150 TB → $0.070/GB | Over 150 TB → $0.050/GB |
| Catalyst | $0 | $0 | $0 | $0 | $0 |
Designed for developers.Trusted by finance team.
For developers
Design the right architecture, not the cheapest one.
- Move data freely between services and clients
- Run AI/ML pipelines without ingress overhead
- Connect integrations without bandwidth math
- Ship faster without billing anxiety
For business and finance
Get predictable invoices as usage grows.
- Enjoy usage-based pricing that scales cleanly
- Set hard spend caps on billable services. Catalyst throttles usage instead of charging overages.
- Stop worrying about hidden bandwidth or per-seat fees
From side projects to data-heavy systems, scale freely.
API-heavy applications
Handle millions of calls without egress overhead.
Event-driven systems
Use high-throughput streams without bandwidth penalties.
AI/ML pipelines
Move training data, embeddings, and model outputs freely.
File-heavy platforms
Upload, process, and serve files without bandwidth fees.
Customer-facing apps
Scale users globally, not your data bill.
Microservices architectures
Inter-service communication stays free at any scale.
Stop paying for moving bits.
Build the architecture you'd build if data were free.
Because on Catalyst, it is.
Build without the bandwidth billFAQs
Catalyst is a cloud platform that charges nothing for data ingress or egress. It's built for developers and engineering teams running workloads where data movement is frequent and high volume: AI pipelines, streaming systems, file platforms, and microservices architectures.
Egress fees charges for data leaving a provider's network function primarily as a lock-in mechanism rather than a reflection of actual infrastructure cost. The expense of moving large datasets away from AWS, GCP, or Azure discourages migration to competing platforms.
Catalyst owns its infrastructure rather than renting capacity from a hyperscaler. Data movement is an operational cost Catalyst absorbs internally. The pricing model is possible because of that infrastructure ownership. It's not a promotional rate or a temporary offer.
Any workload that moves data continuously or at high volume, like machine learning training and inference, real-time event streaming, file delivery, cross-region replication, microservices with heavy inter-service traffic, and consumer apps serving global users.
AWS S3 charges $0.090/GB for the first 10 TB of monthly egress. Google Cloud Storage charges $0.120/GiB. Azure charges $0.087/GB. Catalyst charges $0 at every volume tier.
No. There are no per-seat or per-collaborator fees. You can add engineers, testers, CI systems, and stakeholders without it affecting your bill.