---
title: "Backfill historical blockchain data"
description: "Backfill historical blockchain data into your database or storage with Streams, then keep it flowing as new blocks arrive, with no gaps or duplicates."
canonical: https://www.quicknode.com/solutions/backfill
last-updated: 2026-11-17
---

# Backfill history

**Fill in history with the filters you run live.**

Replay a range of past blocks into your database with the same filters you run live, then keep streaming new blocks as they land.

**URL:** https://www.quicknode.com/solutions/backfill

## Trusted in production

**16 → 91** chains in 12 months. “Quicknode became our primary provider because the consistency held.” [Read the Dune story](https://www.quicknode.com/case-studies/dune)

## Every past block in your database, with the filters you already run.

### One filter, past and present.

Backfills replay past blocks through the same Streams filters you run live, so history and live data always match.

[Streams](https://www.quicknode.com/streams)

### Launch with full history, not just today’s blocks.

Load the range you need into your database, then keep streaming new blocks as they land. No one-off backfill script to maintain.

[One-click Backfills](https://www.quicknode.com/streams/backfills)

### No gaps, no duplicates.

Backfills run in order and pick up where they left off, so your tables match the chain even if a job gets interrupted.

[One-click Backfills](https://www.quicknode.com/streams/backfills)

### Query it where your team already works.

Query it in SQL, call it through an API, or get it delivered to your warehouse on a schedule, with lineage intact end-to-end.

[Analyze onchain data](https://www.quicknode.com/solutions/analyze)

**For AI agents:** Agents use the same data and execution as your team, through MCP, and can pay per request with x402. [How agents use Quicknode](https://www.quicknode.com/agents)

## How Dune grew from 16 to 91 chains.

Dune had no primary provider, so historical and live traces, logs, and receipts were patched by hand. It made Quicknode its primary data source, and went from 16 to 91 chains in 12 months.

> “Quicknode became our primary provider because the consistency held.”
>
> Jonathan Skrzypek, VP of Engineering, Dune

**Built with:** 

[Read the Dune story](https://www.quicknode.com/case-studies/dune)

## Make your first request.

cURL, `Streams API`:

```bash
# Use the API key from your dashboard, then add your filter and destination
curl -X POST https://api.quicknode.com/streams/rest/v1/streams \
  -H "x-api-key: your-api-key" -H "Content-Type: application/json" \
  --data '{"name":"usdc-history","network":"ethereum-mainnet","dataset":"block",
           "start_range":19000000,"end_range":19100000,"destination":"webhook"}'
```

Response, 200 OK:

```
{
  "id": "6f2a…c91d",
  "status": "active",
  "start_range": 19000000
}
```

- [CLI](https://www.quicknode.com/cli): Create endpoints and Streams from your terminal with qn.
- [AI assistant](https://www.quicknode.com/mcp): Add the Quicknode MCP server to Claude or any MCP client.
- [SDK](https://www.quicknode.com/sdk): Typed clients for JavaScript and TypeScript.

**Learn as you build:** [Guides](https://www.quicknode.com/guides/welcome), [Sample apps](https://www.quicknode.com/sample-app-library/welcome), [Courses](https://www.quicknode.com/courses/welcome), [AI Cookbook](https://www.quicknode.com/ai-cookbook)

## Before you start.

### Can this replace my subgraph?

For many apps, yes. Streams and Backfills deliver filtered chain data into your own database, where you shape and query it as you like. If your app reads a GraphQL API over that data, you’d run that layer yourself.

### How far back can I backfill?

As far back as the chain’s history is available on Quicknode, which differs by chain. Check coverage for the chains you need before you scope a backfill.

### How long does a backfill take?

It depends on the range, the chain, and how much your filter keeps. Run a short range first to measure throughput, then size the full job from that.

### Can I query it with SQL instead of writing an indexer?

Yes, for supported chains. SQL Explorer lets you query chain data in SQL, with no indexer to write. See [Analyze onchain data](/solutions/analyze) to start, and check coverage for the chains you need.

### Which chains are covered?

130+ networks across 75+ chains. Coverage varies by product, so check the chains and methods you need.

### What does it cost to start?

The free trial is $0 for one month, with no card needed. Paid plans start at $49 a month. Archive and trace calls use more credits than simple reads. Building something bigger? Talk to our team about a custom solution.

### Why Quicknode?

Because you’re not just getting tools. You’re getting the people behind them. Our engineers have solved problems across trading, stablecoin payments, wallets, fintech, and agentic infrastructure, and that experience grows with every customer. You can bring them in from the first scoping call through your busiest day.

## Building something harder?
Bring in an engineer.

Tell us where it’s stuck, and we’ll help you size it, migrate it, or fix it.

- [Talk to an engineer](https://www.quicknode.com/solutions/backfill#talk)
- [Start building for free](https://www.quicknode.com/signup)
- [Start a backfill](https://www.quicknode.com/solutions/backfill#start)

**Start on your own:** [Read the docs](https://www.quicknode.com/docs/streams/backfilling), [Browse guides](https://www.quicknode.com/guides/tags/streams), [Get the SDK](https://www.quicknode.com/sdk)

## More from Quicknode

- [Quicknode overview](https://www.quicknode.com/index.md)
- [llms.txt index](https://www.quicknode.com/llms.txt)
