In the rapidly evolving world of blockchain technology, one metric stands out as a key indicator of network performance: Transactions Per Second (TPS). This measurement reflects how many transactions a blockchain can process within a single second and serves as a benchmark for scalability, efficiency, and real-world usability.
Understanding TPS is essential for developers, investors, and users alike. It helps assess whether a blockchain can support high-frequency applications like decentralized finance (DeFi), non-fungible tokens (NFTs), or even mainstream payment systems. However, not all TPS figures are created equal. There are three primary types of TPS metrics—each offering unique insights into a network’s capabilities.
The Three Types of TPS Metrics
Real-Time TPS
Real-time TPS represents the actual number of transactions currently being processed on a blockchain per second. Unlike theoretical or peak values, this metric is derived from live network data, making it the most reliable indicator of current performance and user adoption.
Because it reflects real user activity, real-time TPS cannot be manipulated or inflated. A consistently high real-time TPS suggests strong network utilization, while low values may indicate limited usage or bottlenecks in transaction processing.
👉 Discover how leading blockchains maintain high throughput under real-world conditions.
Max Recorded TPS
Max recorded TPS refers to the highest transaction throughput a blockchain has ever achieved in practice. This value typically spikes during periods of intense network activity—such as market volatility, major token launches, or NFT mints—when thousands of users attempt to transact simultaneously.
While this figure demonstrates a network’s ability to handle surges, it doesn’t guarantee consistent performance. Still, a high max recorded TPS indicates robust infrastructure capable of scaling during demand peaks.
Max Theoretical TPS
Max theoretical TPS estimates the upper limit of a blockchain’s transaction capacity under ideal conditions. It's calculated by dividing the maximum block size (in gas or bytes) by the size of the smallest possible transaction—usually a simple native token transfer.
For example:
- Ethereum’s block size limit is approximately 30 million gas.
- A basic ETH transfer consumes 21,000 gas.
- Thus, each block can hold about 1,428 transactions (30M ÷ 21,000).
- With an average block time of 12 seconds, Ethereum’s theoretical maximum is roughly 119 TPS.
However, real-world factors like variable block times, complex smart contracts, and network congestion often reduce actual throughput well below this ceiling.
TPS Comparison Across Leading Blockchains
Below is an updated overview of real-time and peak TPS performance across major blockchains. These figures reflect network behavior as of early 2025 and provide valuable insight into scalability trends shaping the industry.
High-Performance Networks
- Solana: Real-time TPS at 896, with a max recorded TPS of 1,608. Solana's architecture enables extremely fast finality and high throughput, positioning it among the fastest blockchains.
- Hedera: Boasts a real-time TPS of 1,332 and a peak of 3,287, thanks to its hashgraph consensus mechanism that supports enterprise-grade applications.
- Algorand: Achieves 18.02 real-time TPS and has hit a peak of 3,227, showcasing its potential for burst scalability despite lower average usage.
Major Smart Contract Platforms
- BNB Chain: Processes 42.42 transactions per second on average, peaking at 1,731—a testament to its popularity in DeFi and Web3 gaming.
- Polygon: Maintains 44.33 real-time TPS with a max of 282, leveraging layer-2 scaling to enhance Ethereum’s base functionality.
- Ethereum: Currently operates at 11.75 TPS in real time, reaching up to 62.34 during peak loads. Post-upgrade improvements have increased efficiency, but gas fees and congestion remain concerns during spikes.
Emerging and Niche Chains
- Arbitrum & Optimism: As leading Ethereum rollups, Arbitrum records 10.6 real-time TPS (max: 380) while Optimism sits at 4.59 (max: 32.87), reflecting growing adoption of layer-2 solutions.
- opBNB: Designed for high-frequency use cases, opBNB achieves 46.44 real-time TPS and has reached 229, making it ideal for games and social dApps.
- Avalanche & Fantom: Both operate below 5 TPS in daily use but have demonstrated resilience under load, with Avalanche hitting 92.74 and Fantom reaching 181 during stress events.
Lower-Traffic but Functional Networks
Several chains show minimal real-time activity but retain functional capacity:
- Celo: 4.39 TPS (max: 268)
- Gnosis Chain: 2.36 TPS (max: 80.9)
- Polkadot: 0.15 TPS (max: 112)
These networks may serve specialized ecosystems or experience cyclical usage patterns rather than continuous demand.
Frequently Asked Questions (FAQ)
What is a good TPS for a blockchain?
A “good” TPS depends on use case. For general-purpose chains aiming to compete with traditional payment networks like Visa (which handles ~24,000 TPS), higher throughput is critical. Blockchains like Solana and Hedera exceed 1,000 TPS under load, making them competitive for mass adoption. However, for niche or privacy-focused chains, lower TPS with stronger security may be preferable.
Why is there such a gap between theoretical and real-world TPS?
Theoretical TPS assumes optimal conditions—full blocks filled only with simple transactions. In reality, blocks contain diverse operations (smart contracts, NFT mints, cross-chain messages), which consume more resources. Network latency, validator coordination delays, and economic incentives also impact performance.
👉 See how next-gen consensus models are closing the gap between theory and practice.
Does higher TPS always mean better performance?
Not necessarily. While high TPS indicates speed, it must be balanced with decentralization and security—the core tenets of blockchain technology. Some high-throughput chains achieve speed by relying on fewer validators, potentially compromising censorship resistance. True scalability requires harmony between speed, security, and decentralization.
How do layer-2 solutions improve TPS?
Layer-2 protocols (like Arbitrum, Optimism, and Scroll) offload transactions from the main chain (e.g., Ethereum) and batch them for later settlement. This reduces congestion and increases effective throughput without altering the base layer’s security model.
Can blockchains scale infinitely?
No system scales infinitely due to physical and economic constraints—bandwidth limits, hardware requirements, and validator participation costs all impose ceilings. However, innovations like sharding, recursive proofs (ZK-Rollups), and modular architectures are pushing these limits further than ever before.
Is real-time TPS more important than max recorded or theoretical TPS?
Yes—real-time TPS reflects actual user engagement and network health. While max and theoretical values highlight potential, sustained high real-time throughput indicates successful adoption and reliable infrastructure.
Core Keywords
Blockchain scalability, Transactions Per Second (TPS), Real-time TPS, Max Recorded TPS, Max Theoretical TPS, Layer-2 scaling, Network throughput, Blockchain performance