Layer 1 vs Layer 2: What's the Difference?

What a Layer 1 blockchain actually is

A Layer 1, or L1, is a base blockchain that runs entirely on its own. It has its own validators or miners, its own rules for confirming transactions, and its own security that doesn't depend on any other network. Bitcoin is an L1. Ethereum is an L1. Solana is an L1. If you want the full mechanics of how a blockchain works at all, our blockchain explainer covers that from the ground up.

Every L1 is a self-contained system. When you send a transaction on Bitcoin, thousands of independent computers verify and record it directly on Bitcoin's own ledger. Nothing gets outsourced. That independence is exactly what gives an L1 its security, and it's also exactly what starts to cause problems once a lot of people want to use the same chain at the same time.

The problem: you can't maximize everything at once

As L1s like Ethereum grew popular, they ran into what's often called the blockchain trilemma. The idea is simple: a blockchain wants to be decentralized (no single party controls it), secure (hard to attack or corrupt), and scalable (fast and cheap for everyone to use). In practice, it's really hard to max out all three at the same time.

Here's why. A highly decentralized, secure chain needs thousands of independent computers to agree on every single transaction before it's considered final. That agreement process takes time and computing resources. The more you speed it up or cut corners to handle more transactions, the more you tend to concentrate control in fewer hands, or open the door to attacks. Push too hard on scalability, and you usually give up some decentralization or security to get it.

This isn't a design flaw so much as a tradeoff. Ethereum chose to prioritize decentralization and security first, which is part of why, during busy periods, transaction fees on the base chain can spike and confirmations can slow down. That tradeoff is the entire reason Layer 2s exist.

What a Layer 2 is, and how it borrows the L1's security

A Layer 2, or L2, is a separate network built on top of an L1 that handles transactions faster and cheaper, then periodically bundles up what happened and settles it back to the L1. The key part is that last step: because the L2 regularly checks in with the L1, it can inherit the L1's security guarantees without needing to build its own separate army of validators from scratch.

Think of it like a busy local branch that handles most day-to-day activity, while still reporting back to a much more secure central record that everyone trusts. Users get quick, cheap transactions on the L2, and the underlying L1 still acts as the final source of truth if anything is ever disputed.

Rollups: the dominant way L2s work

Most modern L2s are built as rollups. The basic idea is that instead of sending every single transaction to the L1 one by one (slow and expensive), a rollup bundles thousands of transactions together, compresses them, and posts a single compact record back to the L1. That one posting is enough for the L1 to keep the L2 honest, even though the actual transaction activity happened elsewhere.

There are two main flavors of rollup, and the difference comes down to how they prove those bundled transactions are valid. An optimistic rollup assumes transactions are valid by default and gives anyone a window of time (usually about a week) to challenge and prove fraud if something looks wrong. A zero-knowledge rollup, or zk-rollup, instead generates a cryptographic proof alongside the bundle that verifies its validity instantly, without needing a challenge period.

Neither approach is simply "better" across the board. Optimistic rollups tend to be simpler to build and support more general-purpose activity, but they come with that withdrawal delay. Zk-rollups settle faster and don't need a challenge window, but the proof technology is more complex to engineer well.

Real examples you've probably seen

Ethereum is the L1 most L2 activity is built around. Arbitrum and Optimism are two of the largest and most established L2s, both using optimistic rollup designs, and Base (built by Coinbase) is another major one that also runs as an optimistic rollup on top of Ethereum. Solana, by contrast, was designed to handle scale directly on its own L1 rather than leaning on separate L2s, which is one of the bigger architectural differences between the major chains. If you want to see how these chains stack up against each other more broadly, our guide to Ethereum, Solana, Bitcoin and the rest is a good next read.

Why any of this matters if you're just using the thing

Practically speaking, an L2 usually means lower fees and faster transactions than doing the same thing directly on the L1. That's the whole point. But it's not a free lunch: you're accepting slightly different trust assumptions (you're trusting the L2's own operators and its rollup design, on top of trusting the L1 underneath), and if you ever want to move funds back to the L1 from an optimistic rollup, you may run into that withdrawal delay of several days.

None of this makes L2s risky in some alarming way, they're used by millions of people daily and secured by the same L1s backing the assets you already trust. It just means "L1" and "L2" aren't just buzzwords. They describe a real tradeoff between where your transaction actually gets processed and how directly it's backed by the base chain's security.