ClickHouse® is an open source columnar database designed for real-time analytics. Tinybird is a data platform built on top of ClickHouse® that adds developer tooling, TypeScript and Python SDKs, and managed infrastructure so you can ship analytics APIs without managing clusters.
They share the same query engine. The difference is what you do (and don't) manage yourself.
What is ClickHouse®?
ClickHouse® is a column-oriented SQL database open sourced by Yandex in 2016. It powers analytics at scale for companies like Uber, Netflix, and Spotify, excelling at real-time analytical queries.
What is Tinybird?
Founded in 2019, Tinybird maintains its own fork of ClickHouse® and adds the tooling you need to go from data to production API: ingestion, SQL-to-API publishing, SDKs, observability, and CI/CD. The team includes ClickHouse® contributors who regularly ship improvements to the open source project.
Key differences
Infrastructure and database management
Tinybird's approach:
- Provides a higher-level abstraction with out-of-the-box tuning
- Cloud-native architecture with compute-storage separation and zero-copy replication
- Handles complex upgrade processes with daily compatibility checks
- Abstracts away many database configuration details
- Safe schema migrations without downtime
- Intentionally disables a small set of niche SQL functions and table engines for performance and security
ClickHouse® Cloud's approach:
- Exposes more database internals for granular control
- Allows direct ClickHouse® client connections for writing (Tinybird allows reading)
- Offers "Bring Your Own Cloud" (BYOC)
- Provides more control over database tuning
Shared features:
- Multi-cloud availability
- Automatic database updates
- Web-based UI for database interaction
- Managed ETLs and query endpoints
Feature comparison
| Feature | Tinybird | ClickHouse® Cloud |
|---|---|---|
| Uses ClickHouse® as database | ✅ | ✅ |
| Managed service | ✅ | ✅ |
| TypeScript SDK | ✅ | ❌ |
| Python SDK | ✅ | ❌ |
| Native Iceberg table support | ✅ | ✅ |
| Managed Kafka Connector | ✅ | ✅ |
| Managed S3 Connector | ✅ | ✅ |
| Managed GCS Connector | ✅ | ✅ |
| Managed HTTP streaming endpoint | ✅ | ❌ |
| API generation from SQL | ✅ | ✅ |
| Dynamic query parameters in APIs | ✅ | ✅ |
| Auto-generated OpenAPI specs | ✅ | ❌ |
| Secure APIs with custom JWTs | ✅ | ❌ |
| Git integration for CI/CD | ✅ | ❌ |
| Feature branches and local development | ✅ | ❌ |
| Connect ClickHouse® clients | ✅ | ✅ |
| Exposes database settings for tuning | ❌ | ✅ |
| OOTB tuning for high-concurrency queries | ✅ | ❌ |
Performance
Same engine, comparable performance. Tinybird runs ClickHouse® with similar compute resources. The difference is in how the platform tunes for high-concurrency, user-facing workloads out of the box.
Tinybird performance characteristics:
- Shared infrastructure typically achieves 20+ MB/s write throughput and 1K+ QPS with <50ms p95 latency out of the box
- Enterprise plans offer dedicated infrastructure for higher throughput and lower latency
ClickHouse® Cloud performance:
- More direct control over compute sizing
- Performance scales with selected machine configurations
- Pricing aligned with chosen compute resources
Cost analysis
Self-hosted ClickHouse®:
- Lower infrastructure costs but significantly higher development costs
- Requires dedicated engineering resources for cluster management
ClickHouse® Cloud pricing:
- Time-limited free trial available
- Three tiers: Basic, Scale, and Enterprise
- Usage-based billing with per-minute compute metering
- Cost varies by cloud provider, instance size, and active hours
- See ClickHouse® Cloud pricing for current rates
Tinybird pricing:
- Time-unlimited free plan: 10 GB storage, unlimited data processing, 1,000 API requests/day
- Infrastructure-based paid plans with steep volume discounts
- 93% of non-enterprise customers pay less than $100/month
- Median paid Workspace costs less than $10/month
- Enterprise plans offer dedicated infrastructure with usage-based or credit pricing
Developer experience
Fast databases make queries fast. Tinybird makes developers fast.
TypeScript and Python SDKs
Tinybird's SDKs aren't API wrappers. You define datasources, pipes, and endpoints as TypeScript or Python code. The SDK infers types from your schema, so your IDE autocompletes column names and query parameters. Coding agents (Claude Code, Cursor, Copilot) can read and write Tinybird projects because they're just TypeScript or Python.
import { defineEndpoint, node, t, p } from "@tinybirdco/sdk";
export const topPages = defineEndpoint("top_pages", {
params: {
start_date: p.dateTime(),
end_date: p.dateTime(),
limit: p.int32().optional(10),
},
nodes: [
node({
name: "aggregated",
sql: `
SELECT pathname, count() AS views
FROM page_views
WHERE timestamp >= {{DateTime(start_date)}}
AND timestamp <= {{DateTime(end_date)}}
GROUP BY pathname
ORDER BY views DESC
LIMIT {{Int32(limit, 10)}}
`,
}),
],
output: {
pathname: t.string(),
views: t.uint64(),
},
});
Feature branches and local development
Run Tinybird locally with Docker. Create Tinybird branches that map to git branches. Test schema changes and pipeline logic in isolation, then deploy with tb deploy.
Apache Iceberg support
Both Tinybird and ClickHouse® support reading from Apache Iceberg tables. You can query your data lake directly. For sub-100ms API responses, copy frequently accessed data into Tinybird and keep Iceberg as your source of truth.
What Tinybird adds on top of ClickHouse®
- API generation from SQL with built-in documentation, observability, and token management
- Managed HTTP streaming endpoint supporting thousands of requests/second
- Managed Kafka consumer with horizontal scaling
- Managed S3 and GCS connectors to import data from blob storage
- S3 export job queues for long-term storage management
- CLI and SDKs for git-based CI/CD workflows, and to make analytics native to your project in TypeScript or Python
- Feature branches mapped to git branches for isolated development and testing
When to choose each option
Choose self-hosted ClickHouse®:
- Have dedicated data engineering resources
- Want to minimize infrastructure costs
- Need complete architectural control
- Team willing to learn ClickHouse® deeply
Choose ClickHouse® Cloud:
- Prefer more classic ClickHouse® experience on managed infrastructure
- Need granular database configuration control
- Want flexibility to integrate custom tools
- Value direct database access over convenience
Choose Tinybird:
- Want ClickHouse® performance without managing clusters
- Building user-facing analytics and need APIs, not just a database
- Want to define data pipelines in TypeScript or Python
- Need ingestion, API hosting, and observability in one place
Getting started
Tinybird's free plan has no time limit and no credit card. 10 GB storage, unlimited processing, 1,000 API requests/day. Sign up and ship something.
