Arbitrum currently processes approximately 57 transactions per second (TPS) in real-world conditions, with a maximum recorded TPS of 2,036 and a theoretical maximum of 40,000 TPS. As the leading Ethereum Layer 2 scaling solution by ecosystem breadth and stablecoin liquidity, Arbitrum continues to demonstrate significant performance improvements over Ethereum's base layer while maintaining security through optimistic rollup technology — now reinforced by the BoLD permissionless validation protocol.
Understanding Arbitrum's transaction throughput is crucial for developers, investors, and users navigating the Layer 2 ecosystem. Whether you're building DeFi applications, evaluating scaling solutions, or optimizing transaction costs, Arbitrum's performance metrics directly impact your blockchain strategy.
Understanding Arbitrum's Transaction Performance Metrics
Current real-time data shows that Arbitrum operates with impressive efficiency compared to Ethereum's mainnet. The network maintains a block time of just 0.25 seconds, enabling rapid transaction processing while ensuring security through its optimistic rollup architecture — now enhanced by the Stylus upgrade, which introduces WebAssembly execution alongside the EVM.
To understand Arbitrum's performance, we need to examine three key TPS measurements:
Real-time TPS: Current transaction processing rate (~57 TPS)
Maximum recorded TPS: Peak performance under high demand (2,036 TPS)
Theoretical maximum TPS: Technical ceiling under optimal conditions (40,000 TPS)
Eco's cross-chain infrastructure leverages these Layer 2 performance improvements to enable seamless stablecoin transactions across multiple networks, including Arbitrum, without users needing to understand the underlying technical complexity.
Arbitrum vs Ethereum: TPS Performance Comparison
The performance gap between Arbitrum and Ethereum reveals the transformative potential of Layer 2 scaling solutions. Ethereum processes approximately 21.56 TPS on its base layer following the Pectra and Fusaka upgrades in 2025, while Arbitrum achieves significantly higher throughput through its optimistic rollup design.
Performance Benchmarks
Ethereum Mainnet:
Average TPS: 21.56
Maximum recorded TPS: 62.87
Maximum theoretical TPS: 238.1
Block time: 12 seconds
Network capacity expanded significantly through Pectra and Fusaka upgrades
Arbitrum One:
Average TPS: 57
Maximum recorded TPS: 2,036
Theoretical maximum TPS: 40,000
Block time: 0.25 seconds
Average transaction fee: ~$0.004
This represents a 2.6x improvement in real-world performance over Ethereum L1 and dramatically higher potential throughput. The difference becomes even more pronounced during network congestion, when Ethereum fees spike while Arbitrum maintains sub-cent transaction costs.
How Arbitrum Achieves High Transaction Throughput
Arbitrum's impressive TPS performance stems from its sophisticated optimistic rollup architecture, now enhanced by multiple protocol upgrades shipped in 2025. The system processes transactions off-chain while maintaining Ethereum-level security through periodic data posting and fraud-proof mechanisms.
Optimistic Rollup Technology
The "optimistic" approach assumes transactions are valid by default, processing them immediately rather than waiting for cryptographic proof verification. This enables:
Instant transaction finality: Users experience immediate confirmation
Batch processing efficiency: Multiple transactions bundled into single Ethereum submission
Reduced computational overhead: Off-chain execution minimizes mainnet resource usage
BoLD Permissionless Validation
Arbitrum's BoLD (Bounded Liquidity Delay) protocol, deployed on both Arbitrum One and Nova in early 2025, replaced the previous dispute resolution system with permissionless validation. Anyone can now verify transactions and challenge fraud, making the network meaningfully more decentralized. Key improvements include:
Fixed-time dispute resolution windows, eliminating delay attacks
Crowdsourced bonding pools for validator participation
Stronger censorship resistance through open participation
Stylus: Multi-Language Smart Contracts
The Stylus upgrade introduced WebAssembly (WASM) as a co-equal execution environment alongside the EVM, enabling smart contracts written in Rust, C, and C++. This delivers:
10-100x faster execution for compute-heavy operations
Lower gas costs through improved memory efficiency
A broader developer talent pool beyond Solidity
The Eco Protocol builds upon these Layer 2 innovations by providing unified stablecoin balance management across multiple chains, abstracting away the complexity of cross-chain transactions for end users.
Real-World TPS Performance Analysis
Understanding Arbitrum's actual performance requires examining real-world usage patterns rather than just theoretical maximums. Industry analysis reveals that actual demand often differs significantly from technical capacity.
Current Usage Patterns
Arbitrum processes approximately 57 TPS according to live blockchain data, nearly doubling its throughput from 2024. The network has processed over 1.97 billion total transactions since launch, indicating healthy and growing utilization without approaching capacity limits.
Peak Performance Events
During high-demand periods, Arbitrum has demonstrated its ability to scale effectively:
Maximum recorded TPS: 2,036 during volatile market conditions (up from 1,105 in 2024)
Sustained higher throughput during DeFi protocol launches and airdrop events
Minimal fee increases even during peak usage, with average fees holding around $0.004
Demand vs Capacity Analysis
Research shows that most Layer 2 solutions operate well below their theoretical maximums. Arbitrum's current utilization represents approximately 0.14% of its theoretical capacity, providing enormous room for ecosystem growth — even as TVL has surged past $11.65 billion, nearly tripling since mid-2025.
Arbitrum TPS in the Layer 2 Ecosystem Context
Arbitrum's position in the competitive Layer 2 landscape has shifted meaningfully. The L2 market has consolidated, with Arbitrum, Base, and Optimism collectively processing nearly 90% of Layer 2 transactions.
Layer 2 Performance Comparison
Optimistic Rollups:
Arbitrum One: 57 TPS average, 2,036 max recorded
Base: 159.1 TPS average, 1,988 max recorded
Optimism: 21.5 TPS average, 210.1 max recorded
ZK Rollups:
Polygon zkEVM: Variable performance
Starknet: Continued scaling improvements
zkSync Era: ZK-proof execution efficiency
The biggest story here is Base, which grew from ~5 TPS in 2024 to 159 TPS in 2026, driven by Coinbase's 110M+ user funnel. Base announced in February 2026 that it's decoupling from the OP Stack to build its own unified tech stack (base/base), ending its three-year dependency on Optimism.
Arbitrum's Competitive Advantages
Industry experts note that TPS alone doesn't determine rollup success. Arbitrum's advantages include:
EVM + WASM Compatibility: Stylus expands developer reach beyond Solidity
Mature Ecosystem: ~$16B+ TVL with ~37% L2 market share
Stablecoin Liquidity: Deepest stablecoin availability among L2s
Developer Tooling: Complete Ethereum tooling support plus Rust/C/C++ via Stylus
Security Model: Battle-tested optimistic rollups, now with BoLD permissionless validation
Technical Factors Affecting Arbitrum TPS
Several technical elements influence Arbitrum's transaction throughput, from network architecture to external dependencies.
Ethereum Dependency and Fusaka Impact
Arbitrum's TPS depends on Ethereum for data availability and final settlement. The Fusaka upgrade (December 2025) introduced PeerDAS (Peer Data Availability Sampling), which fundamentally changed how nodes confirm blob data from Layer 2s. Impact for Arbitrum:
40-60% reduction in L2 fees within the first month post-Fusaka
Expanded blob throughput enabling higher L2 transaction volumes
Capacity to support 100,000+ TPS across the entire L2 ecosystem
The upcoming Glamsterdam upgrade (planned 2026) will further optimize with enshrined proposer-builder separation (ePBS)
Sequencer Performance
The centralized sequencer currently handles transaction ordering and batch submission. While this creates potential bottlenecks, it also enables:
Predictable transaction ordering
MEV protection for users
Efficient batch optimization
Arbitrum's long-term roadmap includes transitioning to a decentralized fair sequencing model, where a committee of validators collectively determines transaction ordering.
Gas Price Dynamics
Unlike traditional blockchains, where higher fees guarantee faster processing, Arbitrum maintains consistent performance across different fee levels due to its rollup architecture. Post-Fusaka, gas costs have dropped further as blob throughput scaled.
Measuring Beyond TPS: Alternative Performance Metrics
Industry criticism of TPS as a metric has led to alternative measurements that better capture rollup performance complexity.
Gas Per Second (GPS)
Steven Goldfeder from Offchain Labs suggests gas-per-second provides more accurate performance measurement, accounting for transaction complexity rather than a simple count.
User Operations Per Second (UOPS)
With account abstraction now natively supported on Arbitrum (via the ArbOS Callisto upgrade), UOPS metrics better represent actual user interactions, especially for complex DeFi operations using smart accounts and Passkey authentication.
Computational Throughput
Beyond simple transaction counting, measuring actual computational work performed provides insight into network utilization efficiency — particularly relevant now that Stylus enables WASM execution alongside EVM.
Future TPS Improvements and Roadmap
Arbitrum's performance trajectory points toward significant improvements through both protocol upgrades and ecosystem development.
Recent Protocol Upgrades
Multiple upgrades shipped in 2025-2026 have already improved performance:
ArbOS 40 "Callisto" (June 2025): Native account abstraction and Ethereum improvements ahead of schedule
Fusaka Compatibility (December 2025): Ethereum alignment and gas cost optimization
ArbOS Dia (January 2026): Enhanced gas fee predictability, increased chain capacity, and mobile-grade authentication via Passkeys
Arbitrum Nova and AnyTrust
Arbitrum Nova implements AnyTrust technology for applications requiring higher throughput with slightly reduced decentralization assumptions:
Lower fees through off-chain data availability
Higher transaction throughput potential
Suitable for gaming and social applications
Orbit Chains and Scaling
The Orbit framework enables custom Arbitrum chains with configurable parameters:
Custom TPS limits based on application needs
Flexible gas token selection
Independent validation and governance
Ethereum Roadmap Impact
The Glamsterdam upgrade (planned for 2026) will bring enshrined proposer-builder separation to Ethereum, further expanding the data bandwidth available to Layer 2s like Arbitrum. Combined with continued blob throughput scaling, this sets the stage for another step-change in L2 performance.
Optimizing Applications for Arbitrum's TPS
Developers can leverage Arbitrum's expanded performance characteristics to build more efficient applications.
Best Practices for High-Throughput Applications
Transaction Batching:
Group related operations into single transactions
Minimize state changes per transaction
Optimize smart contract execution
Gas Optimization:
Use efficient data structures
Minimize storage operations
Leverage Stylus (WASM) for compute-heavy operations — achieving 10-100x cost reduction vs. EVM for complex logic
User Experience:
Design for instant finality expectations
Implement optimistic UI updates
Leverage native account abstraction for Passkey-based authentication
Handle rare reorg scenarios gracefully
Integration with Cross-Chain Infrastructure
Eco's approach to chain abstraction demonstrates how applications can leverage Arbitrum's performance while maintaining seamless user experiences across multiple networks. This includes:
Unified balance management across chains
Automatic optimization for transaction routing
Simplified developer integration patterns
Practical Implications for Users and Developers
Understanding Arbitrum's TPS performance helps inform practical decisions about blockchain interaction strategies.
For End Users
Transaction Cost Optimization:
Arbitrum provides 90%+ fee savings compared to Ethereum, with average fees around $0.004
Consistent performance regardless of network congestion
Fast confirmation times for improved user experience
DeFi Participation:
Lower barriers to entry for smaller transactions
Efficient yield farming and trading operations
Reduced MEV exposure through sequencer protection
Native account abstraction enables mobile-friendly wallet experiences
For Developers
Application Architecture:
Design for Layer 2-first user experiences
Leverage optimistic finality for responsive interfaces
Choose EVM (Solidity) or WASM (Rust/C/C++ via Stylus) based on performance requirements
Plan for cross-chain functionality from the beginning
Scalability Planning:
Current capacity provides significant growth runway (~0.14% utilization)
Consider Orbit chains for application-specific requirements
Integrate with cross-chain infrastructure for maximum reach
Frequently Asked Questions
What is Arbitrum's current TPS?
Arbitrum currently processes approximately 57 transactions per second in real-world conditions, nearly doubling from ~28 TPS in 2024, with the capability to handle much higher throughput during peak demand periods.
How does Arbitrum's TPS compare to Ethereum?
Arbitrum achieves roughly 2.6x higher TPS than Ethereum's base layer (21.56 TPS post-Pectra/Fusaka upgrades), while offering significantly lower fees (~$0.004 avg) and faster finality times.
What is Arbitrum's maximum theoretical TPS?
Arbitrum's theoretical maximum is 40,000 TPS under optimal conditions, though practical limitations and current demand keep real-world usage much lower.
Why is Arbitrum's actual TPS lower than its theoretical maximum?
Like all blockchains, actual usage depends on network demand, transaction complexity, and external factors like Ethereum's data availability. Current usage represents only about 0.14% of theoretical capacity.
How does Arbitrum maintain security while achieving higher TPS?
Arbitrum uses optimistic rollup technology, processing transactions off-chain while posting transaction data to Ethereum for security. The BoLD protocol enables permissionless fraud-proof validation, allowing anyone to challenge invalid transactions within fixed time windows.
What factors could improve Arbitrum's TPS in the future?
Key improvement vectors include the Glamsterdam Ethereum upgrade (2026), continued blob throughput scaling, sequencer decentralization, Stylus-optimized applications leveraging WASM execution, and application-specific Orbit chains.
How does Arbitrum compare to Base?
Base has surged to ~159 TPS (up from ~5 TPS in 2024), driven by Coinbase's distribution advantage. Arbitrum maintains leadership in ecosystem breadth, stablecoin liquidity, and developer tooling, while Base leads in raw throughput. Both serve different strategic purposes in the L2 landscape.
Is TPS the best metric for evaluating Arbitrum's performance?
While TPS provides useful comparison data, metrics like gas-per-second, user operations per second (especially relevant with native account abstraction), and transaction cost efficiency may better capture real-world performance value.
How does Arbitrum's TPS affect transaction costs?
Higher TPS capability allows Arbitrum to maintain sub-cent fees even during increased usage. Post-Fusaka, average transaction costs have dropped to approximately $0.004 as blob throughput expanded.
Arbitrum's transaction throughput represents a significant advancement in blockchain scalability, offering practical improvements over Ethereum's base layer while maintaining security and decentralization. With current performance of ~57 TPS and theoretical capacity reaching 40,000 TPS, Arbitrum provides substantial headroom for ecosystem growth — now backed by $11.65B+ in TVL and nearly 2 billion processed transactions.
The success of Layer 2 solutions like Arbitrum creates opportunities for innovation in cross-chain infrastructure. Eco's vision of seamless stablecoin transactions across multiple networks, including Arbitrum, demonstrates how TPS improvements translate into better user experiences and expanded blockchain utility. As the ecosystem continues maturing through upgrades like Stylus, BoLD, and Ethereum's Glamsterdam, focusing on practical metrics beyond simple TPS will become increasingly important for evaluating true network performance and user value.