# Xata Pricing Specification

```yaml
version: 1.3
last_updated: 2026-07-01
canonical_html: https://xata.io/pricing
agent_url: https://xata.io/pricing.md
currency: USD
billing_cadence: monthly_in_arrears
```

**For humans:** review this document before sharing with your AI agent. Estimates are for planning only; actual billing follows your Terms of Service or signed agreement.

### How to use this document

1. **Edit [Default assumptions](#default-assumptions)** to match your team.
2. Pick the closest profile in §6 and confirm your inputs with the review checklist.
3. Share the full file (or `https://xata.io/pricing.md`) with your agent.

### Default assumptions

**Edit these values** to match your organization. All §6 worked examples and agent estimates should use these parameters unless you override them explicitly.

#### Global (all Xata Cloud estimates)


| parameter                          | default    | unit         | notes                                                                                                     |
| ---------------------------------- | ---------- | ------------ | --------------------------------------------------------------------------------------------------------- |
| `region`                           | us-east-1  | n/a          | AWS: `us-east-1`, `us-west-2`, `eu-central-1` (1.15×). GCP `us-central1` (1.0×), `asia-southeast1` (1.2×) |
| `plan`                             | xata_cloud | n/a          | Open Source / Xata Cloud / BYOC                                                                           |
| `scale_to_zero`                    | true       | n/a          | Idle branches bill $0/hr compute                                                                          |
| `storage_rate`                     | 0.28       | USD/GB/month | Base rate; × regional multiplier (§4)                                                                     |
| `hours_per_month`                  | 730        | hours        | 24/7 baseline for always-on instances                                                                     |
| `weeks_per_month`                  | 4.33       | weeks        | Branch throughput → monthly                                                                               |
| `branch_delta_overhead_gb`         | 0.1        | GB/branch    | ~100 MB CoW overhead per concurrent branch                                                                |
| `branch_delta_pct_max`             | 0.02       | fraction     | Upper bound only; not used for short-lived PR branches                                                    |
| `concurrent_branches_per_engineer` | 100        | branches     | Planning heuristic; adjust to your observed peak                                                          |
| `avg_branch_lifetime_minutes`      | 20         | minutes      | Typical PR/agent branch active time                                                                       |


#### Example A: Small team, branch-per-PR (SMB)


| parameter                 | default | unit     | notes                                       |
| ------------------------- | ------- | -------- | ------------------------------------------- |
| `engineers`               | 8       | people   | Typical small eng team (range 8–25)         |
| `branches_per_week`       | 2,500   | branches | Observed from high-velocity PR workflows    |
| `avg_concurrent_branches` | 500     | branches | Observed peak; planning cap = 8 × 100 = 800 |
| `database_size_gb`        | 100     | GB       | Production source database                  |
| `instance`                | small   | n/a      | `usd_per_hour` = 0.024 in us-east-1         |
| `staging_env_count`       | 0       | n/a      | Single production clone as branch source    |


#### Example B: Enterprise multi-staging


| parameter                 | default | unit     | notes                                            |
| ------------------------- | ------- | -------- | ------------------------------------------------ |
| `engineers`               | 200     | people   | Typical enterprise eng org (range 100–300)       |
| `branches_per_week`       | 5,000   | branches | High PR volume across teams                      |
| `avg_concurrent_branches` | 2,500   | branches | Observed peak; planning cap = 200 × 100 = 20,000 |
| `database_size_gb`        | 1,000   | GB       | Production source database (1 TB)                |
| `instance`                | medium  | n/a      | `usd_per_hour` = 0.048 in us-east-1              |
| `staging_env_count`       | 4       | envs     | Long-lived staging branch sources                |
| `staging_uptime_fraction` | 0.20    | fraction | Scale-to-zero ~80% of day → active 20% of time   |


### Review checklist (before sending to your agent)

- [ ] Region and cloud (AWS: `us-east-1`, `us-west-2`, `eu-central-1`; GCP `us-central1`, `asia-southeast1`)
- [ ] Instance size for branch sources
- [ ] Production database size (`database_size_gb`)
- [ ] Peak concurrent branches (`avg_concurrent_branches`) for branch-per-PR usage
- [ ] Branch throughput (`branches_per_week`) and average branch lifetime (minutes)
- [ ] Staging environment count and uptime pattern (if applicable)
- [ ] If interested in BYOC → [https://cal.com/team/xata/engineering](https://cal.com/team/xata/engineering)

### Contents

1. [Plan selection](#1-plan-selection)
2. [Constants](#2-constants)
3. [Plans](#3-plans)
4. [Rate card](#4-rate-card)
5. [Cost model](#5-cost-model)
6. [Worked examples](#6-worked-examples)
7. [BYOC](#7-byoc-bring-your-own-cloud)
8. [Agent rules](#8-agent-rules)
9. [FAQ](#9-faq)

---

## 1. Plan selection


| If you need…                                                             | Plan                  | Agent can compute $?                     |
| ------------------------------------------------------------------------ | --------------------- | ---------------------------------------- |
| Self-managed, no platform fee                                            | **Open Source**       | Yes; $0 platform, you pay your own infra |
| Managed branching, fastest start, PR/staging workflows                   | **Xata Cloud (SaaS)** | Yes; see §5 + §6                         |
| Data in your cloud, HIPAA BAA, EA/EDP, dedicated ops, high branch volume | **BYOC**              | **No**; see §7 + discovery call          |


**Links**

- Open Source: [https://github.com/xataio](https://github.com/xataio)
- Xata Cloud signup: [https://console.xata.io/signup](https://console.xata.io/signup)
- Talk to us about BYOC: [https://cal.com/team/xata/engineering](https://cal.com/team/xata/engineering)

---

## 2. Constants


| key                                | value | unit         | notes                                                                     |
| ---------------------------------- | ----- | ------------ | ------------------------------------------------------------------------- |
| `hours_per_month`                  | 730   | hours        | 24/7 monthly compute baseline                                             |
| `weeks_per_month`                  | 4.33  | weeks        | `52 ÷ 12`; for branch-volume math                                         |
| `storage_rate`                     | 0.28  | USD/GB/month | Base rate; × regional multiplier (see §4)                                 |
| `trial_period`                     | 14    | days         | Xata Cloud only                                                           |
| `default_active_hours_per_branch`  | 8     | hours/month  | Legacy default; PR workflows often much lower per branch                  |
| `concurrent_branches_per_engineer` | 100   | branches     | Planning heuristic; actual concurrent varies by workflow                  |
| `branch_delta_overhead_gb`         | 0.1   | GB/branch    | ~100 MB fixed overhead per concurrent branch (typical short-lived branch) |
| `branch_delta_pct_max`             | 0.02  | fraction     | Upper bound: at most 2% of source DB size changed per branch on average   |


**Regions (Xata Cloud)**


| cloud | region            | compute_multiplier | status    |
| ----- | ----------------- | ------------------ | --------- |
| AWS   | `us-east-1`       | 1.0                | available |
| AWS   | `us-west-2`       | 1.0                | available |
| AWS   | `eu-central-1`    | 1.15               | available |
| GCP   | `us-central1`     | 1.0                | available |
| GCP   | `asia-southeast1` | 1.2                | available |


Regions with higher infrastructure costs carry a small regional multiplier, shown with each rate table in §4.

---

## 3. Plans

Boolean cells: `true`, `false`, or `null` (not offered).


| capability          | open_source      | xata_cloud    | byoc                        |
| ------------------- | ---------------- | ------------- | --------------------------- |
| platform_fee_model  | none             | usage_based   | pct_of_managed_cloud_spend  |
| branching           | true             | true          | true                        |
| scale_to_zero       | true             | true          | true                        |
| wake_up_seconds     | 10               | 1             | 1                           |
| storage_engine      | OpenEBS/Mayastor | Xatastor      | Xatastor                    |
| anonymization       | standard         | standard      | advanced_hipaa              |
| provisioning        | self_managed     | fully_managed | customer_cloud_xata_managed |
| support             | community        | standard      | dedicated_24_7              |
| observability       | false            | true          | true                        |
| compliance          | null             | SOC2_GDPR     | SOC2_GDPR_BAA               |
| multi_region_clouds | null             | AWS,GCP       | AWS,Azure,GCP               |


**Price summary:** Open Source is free forever. Xata Cloud starts at $0.012/hr + $0.28/GB/mo storage. Billed by the minute. Only pay for the time storage is in use. The base rate is $0.0000063927 per GB-minute (43,800 minutes per average month). BYOC is quoted after discovery (§7).

**Xata Cloud trial:** Start free with a 14-day Xata Cloud trial.

**Included on Xata Cloud:** CoW branching, scale-to-zero, pgroll, pgstream, observability, standard support. No per-branch, per-developer, or per-database fees. [AWS Marketplace](https://aws.amazon.com/marketplace/pp/prodview-32nlmjyncspls).

---

## 4. Rate card

`usd_per_month = usd_per_hour × hours_per_month` (730).

Rates below apply to **AWS and GCP Xata Cloud**.

### AWS compute rates

#### us-east-1

| instance | vcpus | ram_gib | usd_per_hour | usd_per_month |
| -------- | ----- | ------- | ------------ | ------------- |
| micro    | ≤2    | 1       | 0.012        | 9             |
| small    | ≤2    | 2       | 0.024        | 18            |
| medium   | ≤2    | 4       | 0.048        | 35            |
| large    | 2     | 8       | 0.096        | 70            |
| xlarge   | 4     | 16      | 0.192        | 140           |
| 2xlarge  | 8     | 32      | 0.384        | 280           |
| 4xlarge  | 16    | 64      | 0.768        | 561           |
| 8xlarge  | 32    | 128     | 1.536        | 1121          |

#### us-west-2

Same compute rates as `us-east-1` (1.0× multiplier).

| instance | vcpus | ram_gib | usd_per_hour | usd_per_month |
| -------- | ----- | ------- | ------------ | ------------- |
| micro    | ≤2    | 1       | 0.012        | 9             |
| small    | ≤2    | 2       | 0.024        | 18            |
| medium   | ≤2    | 4       | 0.048        | 35            |
| large    | 2     | 8       | 0.096        | 70            |
| xlarge   | 4     | 16      | 0.192        | 140           |
| 2xlarge  | 8     | 32      | 0.384        | 280           |
| 4xlarge  | 16    | 64      | 0.768        | 561           |
| 8xlarge  | 32    | 128     | 1.536        | 1121          |

#### eu-central-1 (1.15× compute)

| instance | vcpus | ram_gib | usd_per_hour | usd_per_month |
| -------- | ----- | ------- | ------------ | ------------- |
| micro    | ≤2    | 1       | 0.014        | 10            |
| small    | ≤2    | 2       | 0.028        | 20            |
| medium   | ≤2    | 4       | 0.055        | 40            |
| large    | 2     | 8       | 0.110        | 80            |
| xlarge   | 4     | 16      | 0.221        | 161           |
| 2xlarge  | 8     | 32      | 0.442        | 323           |
| 4xlarge  | 16    | 64      | 0.883        | 645           |
| 8xlarge  | 32    | 128     | 1.766        | 1289          |

### GCP compute rates

#### us-central1 (1.0× compute)

Google Cloud `us-central1` (Iowa) is available on Xata Cloud. Rates match AWS `us-east-1`.

| instance | vcpus | ram_gib | usd_per_hour | usd_per_month |
| -------- | ----- | ------- | ------------ | ------------- |
| micro    | ≤2    | 1       | 0.012        | 9             |
| small    | ≤2    | 2       | 0.024        | 18            |
| medium   | ≤2    | 4       | 0.048        | 35            |
| large    | 2     | 8       | 0.096        | 70            |
| xlarge   | 4     | 16      | 0.192        | 140           |
| 2xlarge  | 8     | 32      | 0.384        | 280           |
| 4xlarge  | 16    | 64      | 0.768        | 561           |
| 8xlarge  | 32    | 128     | 1.536        | 1121          |

#### asia-southeast1 (1.2× compute)

Google Cloud `asia-southeast1` (Singapore) is available on Xata Cloud. Rates are 20% above `us-east-1`.

| instance | vcpus | ram_gib | usd_per_hour | usd_per_month |
| -------- | ----- | ------- | ------------ | ------------- |
| micro    | ≤2    | 1       | 0.014        | 10            |
| small    | ≤2    | 2       | 0.029        | 21            |
| medium   | ≤2    | 4       | 0.058        | 42            |
| large    | 2     | 8       | 0.115        | 84            |
| xlarge   | 4     | 16      | 0.230        | 168           |
| 2xlarge  | 8     | 32      | 0.461        | 337           |
| 4xlarge  | 16    | 64      | 0.922        | 673           |
| 8xlarge  | 32    | 128     | 1.843        | 1345          |

### Storage rates (all regions)

Xatastor NVMe. Storage uses the **same regional multiplier as compute**.

| region            | storage_rate (USD/GB/month) |
| ----------------- | --------------------------- |
| `us-east-1`       | 0.28                        |
| `us-west-2`       | 0.28                        |
| `eu-central-1`    | 0.32                        |
| `us-central1`     | 0.28                        |
| `asia-southeast1` | 0.34                        |

---

## 5. Cost model

### Inputs & derived values

```
branches_per_month = branches_per_week × weeks_per_month

active_hours_per_branch = avg_branch_lifetime_minutes ÷ 60

concurrent_branches_planning = engineers × concurrent_branches_per_engineer

delta_gb_per_branch = branch_delta_overhead_gb

delta_gb_per_branch_upper = max(branch_delta_overhead_gb, database_size_gb × branch_delta_pct_max)
```

### Line items (Xata Cloud)

```
production_clone_compute = instance_hourly_rate × hours_per_month

production_clone_storage = database_size_gb × storage_rate

staging_compute = instance_hourly_rate × active_hours_per_month × staging_env_count
  where active_hours_per_month = hours_per_month × uptime_fraction

branch_compute = instance_hourly_rate × active_hours_per_branch × branches_per_month

branch_delta_storage = avg_concurrent_branches × branch_delta_overhead_gb × storage_rate

branch_delta_storage_upper = avg_concurrent_branches × max(branch_delta_overhead_gb, database_size_gb × branch_delta_pct_max) × storage_rate
```

---

## 6. Worked examples

Both examples use parameters from [Default assumptions](#default-assumptions). Math uses §2 Constants and §4 Rate card.

---

### Example A: Small team, branch-per-PR

**Segment:** SMB; high branch velocity with many concurrent scaled-to-zero branches.

**Parameters:** Example A row in [Default assumptions](#default-assumptions) (8 engineers, 2,500 branches/week, 500 concurrent, 100 GB, `small` instance).

**Derived values**


| Derived                        | Calculation               | Value          |
| ------------------------------ | ------------------------- | -------------- |
| `branches_per_month`           | 2,500 × 4.33              | 10,825         |
| `active_hours_per_branch`      | 20 ÷ 60                   | 0.333          |
| `concurrent_planning_estimate` | 8 × 100                   | 800            |
| `delta_gb_per_branch`          | 0.1 GB overhead (typical) | 0.1            |
| `byoc_tier_fit`                | 500 concurrent            | Team (≤ 1,000) |


**Monthly estimate**


| line_item                | calculation            | usd/month |
| ------------------------ | ---------------------- | --------- |
| production_clone_compute | 0.024 × 730            | 17.52     |
| production_clone_storage | 100 × 0.28             | 28.00     |
| branch_compute           | 0.024 × 0.333 × 10,825 | 86.58     |
| branch_delta_storage     | 500 × 0.1 × 0.28       | 14.00     |
| **total**                |                        | **~146**  |


**Upper-bound check** (2% delta): 500 × max(0.1, 100 × 0.02) × 0.28 = 500 × 2 × 0.28 = **$280**. Still far below full copy cost.

**Takeaway**

- **Total:** ~$150/month
- **Dominant cost:** branch compute (~$87)
- **Storage delta:** ~$14

---

### Example B: Enterprise staging + branches

**Segment:** Enterprise; multiple staging sources with high concurrent branch count.

**Parameters:** Example B row in [Default assumptions](#default-assumptions) (200 engineers, 5,000 branches/week, 2,500 concurrent, 1 TB, 4 staging envs @ 20% uptime, `medium` instance).

**Derived values**


| Derived                          | Calculation               | Value                  |
| -------------------------------- | ------------------------- | ---------------------- |
| `branches_per_month`             | 5,000 × 4.33              | 21,650                 |
| `active_hours_per_branch`        | 20 ÷ 60                   | 0.333                  |
| `concurrent_planning_estimate`   | 200 × 100                 | 20,000                 |
| `staging_active_hours_per_month` | 730 × 0.20                | 146                    |
| `delta_gb_per_branch`            | 0.1 GB overhead (typical) | 0.1                    |
| `byoc_tier_fit`                  | 2,500 concurrent          | Company (1,000–10,000) |


**Monthly estimate**


| line_item                | calculation            | usd/month |
| ------------------------ | ---------------------- | --------- |
| production_clone_compute | 0.048 × 730            | 35.04     |
| production_clone_storage | 1,000 × 0.28           | 280.00    |
| staging_compute          | 0.048 × 146 × 4        | 28.03     |
| branch_compute           | 0.048 × 0.333 × 21,650 | 346.72    |
| branch_delta_storage     | 2,500 × 0.1 × 0.28     | 70.00     |
| **total**                |                        | **~760**  |


**Upper-bound check** (2% delta): 2,500 × max(0.1, 1,000 × 0.02) × 0.28 = 2,500 × 20 × 0.28 = **$14,000**. Use 100 MB/branch default for short-lived PR branches and reserve 2% for long-running or write-heavy branches.

**Takeaway**

- **Total:** ~$760/month
- **Dominant cost:** branch compute (~$347)
- **Storage delta:** ~$70
- **BYOC fit:** Company tier

---

## 7. BYOC (Bring Your Own Cloud)

### Model

- Xata runs in **your** cloud account (AWS, Azure, GCP).
- You pay your cloud provider for Xata-managed compute and storage.
- Xata charges a **management fee**, a percentage of your **Xata-managed cloud spend**. The exact percentage is confirmed in your proposal after a discovery call and isn't listed in this document.

### Minimum annual commitment

BYOC contracts include a **minimum annual commitment** tied to your capacity tier (concurrent branch range). Specific amounts are in your order form, not in this public spec.

### Estimate process

Xata produces a **written pricing estimate** once there is enough detail on regions, scale, compliance, topology, and branch patterns (similar to §6). That typically requires a short discovery call.

**Book discovery:** [https://cal.com/team/xata/engineering](https://cal.com/team/xata/engineering)

### Capacity tiers (fit guide)


| tier       | concurrent_branches_max | typical_segment                                             |
| ---------- | ----------------------- | ----------------------------------------------------------- |
| Team       | 1,000                   | Engineering orgs replacing shared staging (Example A scale) |
| Company    | 10,000                  | Company-wide rollouts, multi-staging (Example B scale)      |
| Platform   | 100,000                 | Platforms provisioning DBs per customer                     |
| Hyperscale | 100,000+                | Massive multi-tenant provisioning                           |


Tier selection depends on **peak concurrent branches** and architecture, not engineer headcount alone.

### What the management fee covers

CoW branching platform, scale-to-zero orchestration, 24/7 operational management, cluster management, upgrades/patches, HIPAA BAA where applicable.

### Cloud commitments

BYOC spend in your account counts toward AWS/GCP/Azure enterprise agreements. Cloud credits apply to your provider bill; the management fee is generally calculated on pre-credit Xata-managed spend unless your contract states otherwise.

---

## 8. Agent rules

### Do

- Use [Default assumptions](#default-assumptions) and §2–§6 for **Xata Cloud** estimates.
- Ask for missing inputs: plan → cloud → region → instance → `database_size_gb` → `avg_concurrent_branches` → `branches_per_week` → `avg_branch_lifetime_minutes` → staging env count/uptime.
- For AWS and GCP regions, use §4 rate tables.
- Use **Example A** or **Example B** as sanity checks for similar profiles.
- For BYOC, explain the model and link [https://cal.com/team/xata/engineering](https://cal.com/team/xata/engineering).

### Do not

- State BYOC **management fee percentages** or **minimum commitment dollar amounts** from this doc.
- Apply 60–90% savings as a numeric multiplier.
- Merge cloud provider spend and Xata fees without labeling each.

### Recommended output format

1. **Assumptions** (inputs table)
2. **Derived values** (concurrent branches, branches/month)
3. **Line items** (USD/month each)
4. **Total** (USD/month)
5. **Caveats** (short-lived branches, scale-to-zero, excludes BYOC unless scoped)
6. **Closest example** (A or B)

---

## 9. FAQ

### Do you charge per branch?

No. Xata Cloud is usage-based on compute and storage time. BYOC uses capacity tiers by concurrent branch range. See §3 and §5.

### Is there a free tier?

No. Our core platform is available under Apache 2.0. Xata Cloud starts with a 14-day free trial.

### Can we start on SaaS and move to BYOC?

Yes. Same Postgres, branching, and workflows. Teams typically move for scale, compliance, data residency, or enterprise cloud commitments. See §7.