Polygon: EVM PoS Sidechain with Heimdall v2 Finality
Polygon (POL, formerly MATIC) is an EVM-compatible Proof-of-Stake sidechain launched May 2020. It pairs Bor block producers with a Heimdall v2 / CometBFT layer that checkpoints to Ethereum, delivering ~2 s block times and roughly 5 s deterministic finality. With $1.04B TVL and sub-cent gas, it sits alongside Arbitrum, Base, and Optimism in the L2/scaling category — but as a sovereign sidechain, not a rollup. POL launched in September 2024 as the new staking and gas asset for the AggLayer-aligned Polygon ecosystem.
Polygon serves two distinct constituencies. For high-frequency consumer apps — prediction markets (Polymarket alone holds $415M), payments, RWA platforms, NFT mints — it offers gas costs that EVM mainnet simply cannot match, paired with the same MetaMask/Rabby/Ledger tooling developers already know. For protocol teams, it is one of the most battle-tested EVM execution environments outside Ethereum L1, with five years of uptime, deep CEX support for POL/MATIC, and an explorer (polygonscan.com) that mirrors Etherscan UX. The 2025 Heimdall v2 hard fork cut probabilistic confirmation from minutes to a deterministic ~5 s window via CometBFT, closing the UX gap with newer high-speed L1s. The trade-off is honest: Polygon is a sidechain checkpointed to Ethereum, not a rollup — its security model relies on its own validator set plus periodic L1 anchoring, distinct from Arbitrum or Base where state validity inherits Ethereum directly. For non-custodial cross-chain swap flows, this profile (cheap gas + 5 s finality + $882M circulating USDT) makes Polygon a workhorse settlement layer for stablecoin and consumer traffic, while developers reaching for trust-minimised rollup guarantees still gravitate to OP Stack or Arbitrum Nitro.
About Polygon
Polygon mainnet launched on 2020-05-30, originally branded Matic Network and positioned as an Ethereum scaling sidechain rather than a Layer-2 rollup. The architecture splits responsibilities across two layers: Bor, a Geth-fork producing ~2 s EVM blocks, and Heimdall, a CometBFT (formerly Tendermint) validator layer that selects block producers and periodically checkpoints state to Ethereum L1. This sidechain design — sovereign consensus + L1 anchoring — predates and is architecturally distinct from optimistic and ZK rollups, which is why honest classifications place Polygon PoS in its own category even when product UX lines up with L2s.
The consensus stack is Proof-of-Stake. Validators stake POL (the upgraded successor to MATIC) on Ethereum, are selected into the Heimdall set, and produce Bor blocks in rotation. Heimdall v2, activated in the 2025-07-10 hard fork, replaced the original probabilistic finality model with CometBFT-style deterministic finality at roughly 5 s — a meaningful UX upgrade for users previously waiting through reorg risk. Theoretical throughput tops out near 1,000 TPS; sustained typical throughput is closer to 60 TPS, reflecting real workload rather than benchmark conditions.
The economic redesign is equally consequential. On 2024-09-04 the network began migrating MATIC to POL, a hyperproductive token engineered to secure not only Polygon PoS but the broader AggLayer constellation (zkEVM, CDK chains, future restaked chains). POL functions as gas, staking collateral, and governance asset. The token swap is 1:1, but the underlying contract is new, and CEX/DEX listings have rotated accordingly — a detail that matters for non-custodial cross-chain swap routing that needs canonical contract addresses.
Polygon's native on-chain DeFi TVL stands at $1.04B today, led by Polymarket International ($415M in prediction-market liquidity), with a long tail across lending (Compound Blue, $3M), RWA (Binaryx, $8.65M), yield (Sphere Finance, $6.2M), and DEXs (W-DEX, RamsesX, GT3). Tracked separately, the Polygon Bridge contract on Ethereum L1 holds $2.24B in escrowed assets bound for Polygon — a figure that sits outside the chain's native DeFi TVL but illustrates how much cumulative bridged value gravitates here. The chain's reported native-stablecoin balance is $0; virtually all dollar liquidity arrives as bridged USDT, currently $882M circulating, making Polygon the 7th-largest USDT venue globally.
Polygon technical parameters
Polygon's technical identity is shaped by a deliberate split-layer design that few peers replicate. Understanding it requires separating Bor (the execution layer everyone interacts with) from Heimdall (the validator and checkpointing layer that finalises state). The combination is what allows POL to keep EVM-equivalent semantics while running an independent validator set and posting compact checkpoints to Ethereum on a regular cadence.
| Consensus | PoS (Bor block producers + Heimdall v2 / CometBFT checkpointing) |
|---|---|
| VM | EVM |
| Block time | 2 s |
| Finality | 5 s |
| TPS | 60 typical / 1k max |
| Gas token | POL (formerly MATIC) |
| Launched | 2020-05-30 |
| Token standard | ERC-20 / ERC-721 / ERC-1155 |
| Address | hex (0x-prefixed, EVM) |
Consensus mechanism
Consensus on Polygon runs in two coordinated tiers. Heimdall is a CometBFT-based PoS network where validators stake POL (the contract lives on Ethereum L1) and reach BFT agreement on validator-set changes, checkpoints, and finality. Heimdall periodically writes a Merkle root of recent Bor blocks back to a checkpoint contract on Ethereum, giving Polygon its anchoring guarantee. Bor is the execution layer — a Geth fork producing EVM blocks every ~2 seconds. Block production is delegated by Heimdall to a rotating producer set, so Bor itself is not a separate consensus protocol; it is a sequencing layer driven by the BFT layer above. Before the 2025-07-10 Heimdall v2 upgrade, users effectively waited for the next Ethereum checkpoint (minutes) for hard finality. Heimdall v2 introduces CometBFT-style deterministic finality at the Heimdall layer, collapsing wait time to roughly 5 s for an irreversible state — a step change for swap UX, payment confirmation, and intent-based settlement.
Performance context
Polygon's published numbers — ~2 s block time, ~5 s finality, ~60 TPS typical, ~1,000 TPS theoretical max — only make sense in peer context. Ethereum L1 finalises in ~768 s and sustains ~15 TPS; Polygon is roughly 150x faster to finality and 4x higher throughput in everyday conditions. Against high-speed L1s like Solana (12.8 s finality, ~3,000 TPS typical) or Aptos (0.65 s, ~800 TPS), Polygon is slower per-transaction but vastly more EVM-native — the trade-off is tooling parity, not raw speed. Versus its EVM L2 peers, Polygon's 5 s finality is dramatically faster than Arbitrum's 7-day optimistic challenge window or Base's ~780 s — though only because Polygon's finality is sovereign, not rollup-inherited.
Polygon ecosystem map
Polygon's ecosystem reflects its core thesis — high-throughput, low-fee EVM execution for consumer-scale apps. Unlike Ethereum L1 where DeFi blue chips dominate TVL, Polygon's top protocols span prediction markets, RWAs, lending, and DEX liquidity, with bridged value from Ethereum providing the gravitational centre.
Bridge
Polygon Bridge holds $2.24B locked from Ethereum L1 — the canonical lock-and-mint route for ERC-20s entering the chain. This figure is tracked separately from Polygon's $1.04B native DeFi TVL and underlines how much of the chain's economic activity originates as bridged Ethereum-side liquidity.
Prediction Market
Polymarket International commands $415M TVL on Polygon, making prediction markets the single largest non-bridge category. The pairing is deliberate: sub-cent gas plus 5 s finality is uniquely suited to high-frequency micro-bets that would be uneconomical on L1.
RWA
Binaryx Platform ($8.65M) anchors a small but growing tokenised real-world-asset segment. Polygon has been a deliberate target for institutional RWA pilots — KYC-gated, regulator-friendly venues — because of its low gas profile and mature compliance tooling integrations.
Lending
Compound Blue ($3M) represents a focused lending presence rather than the dominant Aave/Compound deployments that historically held nine-figure TVL on Polygon. The current footprint reflects a multi-cycle migration of money-market liquidity toward Base and Arbitrum.
DEX
On-chain DEX liquidity is fragmented across RamsesX Polygon CL ($537K), W-DEX ($2.07M), and GT3 ($507K). The long-tail spread — rather than a single dominant venue — means cross-chain swap routers must aggregate multiple Polygon-native pools to surface competitive pricing.
Yield
Sphere Finance ($6.2M) headlines the yield/strategy vault category, providing automated allocation across Polygon-native DEXs and lending markets. Yield infrastructure on Polygon trails the Arbitrum and Base equivalents in TVL but benefits from significantly lower rebalancing gas costs.
| # | Protocol | Category | TVL |
|---|---|---|---|
| 1 | Polygon Bridge | Chain | $2.24B |
| 2 | Polymarket International | Prediction Market | $415.26M |
| 3 | Binaryx Platform | RWA | $8.65M |
| 4 | Sphere Finance | Yield | $6.20M |
| 5 | Compound Blue | Lending | $3.00M |
| 6 | W-DEX | Dexs | $2.07M |
| 7 | RamsesX Polygon CL | Dexs | $536.54K |
| 8 | GT3 | Dexs | $507.10K |
Polygon vs peers
Within the EVM L2 cohort, Polygon's profile is the most idiosyncratic — it predates the rollup era, runs sovereign PoS consensus, and trades rollup security inheritance for dramatically faster finality. The comparison below uses live data from peer chains in the same category.
| Chain | Block | Finality | TPS | TVL |
|---|---|---|---|---|
| Arbitrum | 250 ms | 168 h | 40 | $1.24B |
| Optimism | 2 s | 168 h | 30 | $0.00 |
| Base | 2 s | 13 min | 1.5k | $3.84B |
| Polygoncurrent | 2 s | 5 s | 60 | $1.05B |
| Starknet | 2.5 s | 1.5 h | 3 | $184.41M |
| X Layer | 2 s | — | 100 | $73.95M |
| Scroll | 3 s | 1 h | 30 | $10.79M |
| Abstract | 1 s | — | — | $11.13M |
Comparison insights
- TVL ranking: Among EVM L2s in the dataset, Base leads at $3.84B, Arbitrum at $1.24B, Polygon at $1.04B, Starknet at $184M, X Layer at $74M, and Scroll at $10.8M. Polygon holds the third-largest TVL in the category — competitive given that Base and Arbitrum benefit from being canonical OP Stack / Nitro rollups inheriting Ethereum security, while Polygon's TVL is built on a sovereign sidechain model.
- Finality, honestly stated: Polygon's ~5 s deterministic finality (post Heimdall v2) is the fastest in the EVM L2 cohort. Arbitrum and Optimism require ~604,800 s (7 days) for unconditional withdrawal finality due to the optimistic challenge window; Base inherits OP Stack's ~780 s checkpoint cadence; Starknet's STARK proofs settle in ~5,400 s; Scroll in ~3,600 s. The asterisk: Polygon's finality is its own validator set's BFT agreement, not Ethereum-inherited. Different threat models, different guarantees.
- Throughput per dollar of TVL: Polygon's ~60 typical TPS against $1.04B TVL puts utilisation density between Arbitrum (~40 TPS, $1.24B TVL) and Base (~1,500 TPS, $3.84B TVL). Base's reported TPS reflects its position as Coinbase's consumer on-ramp; Polygon's lower figure reflects a more diversified workload mix without a single dominant traffic source.
- Architecture honesty: Of the eight EVM L2s in the dataset (arb, op, base, pol, starknet, xlayer, scroll, abs), only Polygon is not a rollup. Arbitrum and Optimism are optimistic rollups; Base is OP Stack; Starknet and Scroll are validity (ZK) rollups; X Layer is a Polygon CDK validium; Abstract is a ZKsync ZK Stack rollup. Polygon PoS is a PoS sidechain with checkpointing — categorisation parity hides genuine security-model differences.
- Stablecoin density: Polygon hosts $882M circulating USDT (0.47% of global USDT supply), the 7th-largest venue. Arbitrum, the closest L2 comparable in the USDT data, hosts $1.02B (0.54%). Despite Arbitrum's larger TVL, Polygon punches above its weight on stablecoin circulation per TVL dollar — a function of Polygon's payments and consumer-app workload mix.
Polygon timeline
Polygon's history maps the evolution of Ethereum scaling itself. Mainnet launched on 2020-05-30 as Matic Network, a Plasma + PoS sidechain combination targeting cheap EVM execution. Through 2021-2022 the network became the default home for consumer dApps and NFT mints priced out of Ethereum L1, peaking at multi-billion-dollar TVL during the bull cycle. The team rebranded as Polygon Labs and expanded the roadmap to include Polygon zkEVM, Polygon Miden, and the Chain Development Kit (CDK), pivoting from a single-chain narrative to a multi-chain L2 vision. On 2024-09-04 the MATIC-to-POL migration began — a 1:1 token swap underpinning a new hyperproductive token model designed to secure the AggLayer constellation of Polygon-aligned chains, not just Polygon PoS. The most consequential technical event arrived on 2025-07-10 with the Heimdall v2 hard fork, which replaced the original Tendermint stack with CometBFT and introduced deterministic ~5 s finality, closing a long-standing UX gap versus newer high-speed L1s. Polygon PoS has not suffered a consensus-halting outage of the scale that affected Solana in 2022-2023, nor a bridge exploit comparable to BSC's 2022 cross-chain bridge incident, though the broader Polygon ecosystem has faced routine smart-contract incidents at the application layer (as has every major EVM chain). The trajectory is recognisably one of incremental, in-production hardening rather than disruptive reinvention.
- 2020-05-30launchPolygon PoS mainnet launch
- 2024-09-04milestoneMATIC → POL token migration begins
- 2025-07-10upgradeHeimdall v2 hard fork — ~5s finality via CometBFT
Developer reference
Polygon is one of the most developer-friendly EVM environments outside L1 — tooling parity with Ethereum is near-complete. The official RPC endpoint is https://polygon-rpc.com; the canonical block explorer is polygonscan.com (Etherscan-compatible UI and API). Addresses are standard hex-encoded 0x-prefixed EVM addresses, identical in format to Ethereum mainnet, so existing key material, MetaMask accounts, and Ledger derivation paths port directly. Token standards are ERC-20, ERC-721, and ERC-1155 — Solidity contracts compile and deploy with Hardhat, Foundry, or Truffle without modification, and viem/ethers.js work out of the box. Wallet support spans MetaMask, WalletConnect, Rabby, Coinbase Wallet, and Ledger. Official documentation lives at https://docs.polygon.technology. The native gas and staking asset is POL (chain ID 137); developers building integrations should target POL-aware contract addresses rather than the legacy MATIC contract.
Assets swappable on Polygon
Grouped by category. Click any asset to open its swap page for a live quote.
Majors
1 assets
DeFi assets
1 assets
Polygon settle-time comparison
Shorter bars mean faster confirmations. Real settle time also depends on network congestion — figures are indicative.
Polygon asset coverage comparison
Longer bars mean more assets are swappable on that chain.
Popular swap routes involving Polygon
Routes below reflect actual user preference. Click to jump straight to the swap page.
Polygon FAQ
01Is Polygon a Layer-2 or a sidechain?
Technically a PoS sidechain, not a rollup. Polygon PoS runs its own validator set (Heimdall v2 / CometBFT) producing Bor blocks every ~2 s, with periodic checkpoints written back to Ethereum L1. True L2 rollups (Arbitrum, Optimism, Base, Starknet, Scroll) inherit Ethereum's security via fraud or validity proofs; Polygon PoS relies primarily on its own staked validator set plus L1 anchoring. The distinction matters for trust assumptions even when UX is similar.
02What is Polygon's finality time?
Approximately 5 seconds since the 2025-07-10 Heimdall v2 hard fork. Heimdall v2 introduced CometBFT-style deterministic finality, replacing the previous probabilistic model that required waiting for the next L1 checkpoint (minutes) for hard finality. Block time on the Bor execution layer is ~2 seconds; deterministic finality from Heimdall lands at ~5 s.
03How decentralised is Polygon?
Polygon PoS uses a Heimdall validator set of ~100 active validators staking POL on Ethereum L1. Block production rotates among elected producers selected from this set. The validator count is larger and more permissionless than rollup sequencer arrangements (most rollups still run a single sequencer in production), but smaller than Ethereum L1's ~1M+ validator beacon chain. Decentralisation is real and quantifiable, not theatre — but it is its own validator set, not Ethereum's.
04What's the difference between MATIC and POL?
POL is the upgraded successor to MATIC, with migration launched on 2024-09-04. The swap is 1:1 in value, but POL is a new ERC-20 contract designed as a 'hyperproductive' token securing the broader AggLayer ecosystem of Polygon-aligned chains, not only Polygon PoS. POL is the chain's gas asset, staking collateral, and governance token. Legacy MATIC holdings can be migrated through official upgrade contracts; developers should target POL contract addresses for new integrations.
05What is Polygon's TVL and main usage?
Polygon's native on-chain DeFi TVL stands at $1.04B as of 2026-06-06, led by Polymarket International ($415M in prediction-market liquidity). Separately, the Polygon Bridge contract on Ethereum holds $2.24B in escrowed assets bound for Polygon — tracked outside the chain's native TVL but indicative of cumulative bridged value. Polygon also hosts $882M of circulating USDT — the 7th-largest USDT venue globally — making it a workhorse for consumer payments, prediction markets, RWA pilots, and stablecoin movement rather than a deep-DeFi yield venue.