Crypto earns its keep when people need one shared record and no single operator running it. Kaspa's bet is that Bitcoin-style Proof-of-Work security and censorship resistance can run closer to real time without giving up either one.
Five jobs, and Kaspa is at a different stage on each one.
| Crypto job | Kaspa angle | Status |
|---|---|---|
| Self-custodied money | Fair-launched Proof-of-Work UTXO asset with direct wallet control. | Live |
| Fast payment feel | 10 BPS blockDAG gets a payment into a block faster than slow PoW chains. | Live, with confirmation tradeoffs |
| One shared record | GHOSTDAG orders parallel honest blocks in a DAG. | Live |
| Assets with rules | Toccata put spend rules, ZK hooks, sequencing commitments, and vProgs groundwork into L1. | Covenant rules live on L1; wallet and app evidence next |
| Group commitments | Base RTD today. RTD-derived attestations, oracles, and TangVM come later. | Base idea live; app systems still future work |
Six families
Sort a ranking page by market cap and BTC, ETH, USDT, DOGE, and the week's meme token all land in one column, as though they were the same kind of thing. They were never chasing the same job.
Bitcoin has the longest track record as a store of value. Ethereum has the deepest pool of smart contracts. Solana has the most polished fast-app feel. Kaspa leads on none of the three today. Ranking them is a tool job rather than a prose job, so the chain comparer scores twenty layer ones on fields measured the same day and reorders as the dials move.
A block every ten minutes is the price of settlement money that holds still. Kaspa keeps the Proof of Work and the fixed supply, then swaps the single chain for a blockDAG that keeps the blocks other PoW chains discard as orphans.
Rollups fix throughput and split liquidity, each with its own bridge, sequencer, and trust to check. Toccata put Kaspa's app rules on the base layer instead, settling against the same order every payment uses.
Users, developers, and liquidity all sit in one place, with no bridges to spread across. Cardano, Avalanche, Sui, TON, Hedera, and Polkadot started from the app side too.
Issuers, reserves, bank access, and the promise to redeem sit behind every dollar token. TRON carries most of that traffic on delegated proof of stake, where 27 elected super representatives take turns making blocks.
BNB's strength and its risk are the same fact: it works because Binance works. XRP's servers agree using validator lists they each choose to trust, which is neither proof of work nor proof of stake. Kaspa is fair-launched PoW with no company behind it.
Culture is the product, and neither competes on design. Dogecoin adds proof-of-work roots and odd staying power.
The claim here is not that Kaspa beats any of the six. It is that the combination is rare: mined issuance, fixed supply, blocks that arrive in parallel, a payment feel you would use daily, and app rules that stay on one layer.
Pick your constraint
No crypto design gets speed, security, privacy, and a network nobody controls. Push on one and the cost lands somewhere else.
| Goal | Why people want it | What it costs |
|---|---|---|
| Decentralization | No easy single control point. | Harder upgrades, worse UX, slower agreement, more on you. |
| Security | Hard-to-fake history and valid supply. | Fees, power bills, staked capital, and a security budget that holds up. |
| Speed | Fast inclusion and better user feedback. | Blocks clash more, nodes get heavier, checks thin out, control narrows. |
| Privacy | Less watching. Every coin spends the same. | Heavier math, bigger data, harder to audit, fewer exchanges. |
| Self-custody | You hold the keys. No company account. | No password reset, no chargeback, no support line. |
Visa settles fast because nobody outside Visa has to check its work. Banks, contracts, one company's servers, and a fraud team do it, and that company can undo errors, block users, and change the rules at will. A public chain asks thousands of strangers to check the same rules, with no one in charge, so independent nodes can throw out bad history without asking anyone. That is the slow part. The app layer can be Visa-fast; a base layer anyone can check cannot.
Concentration arrives by a different door in each model. ASIC cost gathers around whoever has the capital, the cheap power, and the chips. Proof of stake skips the hardware, so holding the token buys the say instead.
Every one of those costs still lands on Kaspa. The bet it makes against them is narrow, and it starts with keeping the blocks that lose the race.
Bitcoin proved strangers can agree on scarce digital money with nobody in charge. It pays for that with a slow block interval, picked partly because network lag has to stay small next to block production. Every Nakamoto-style system carries that tradeoff by design.
Kaspa comes out of the research line asking whether mining work that a single chain throws away as a collision can count toward consensus instead.
The goal has a name. RTD, Real-Time Decentralization, is Hashdag's term for a partially synchronous system that moves as fast as the network really allows: fast when conditions are clean, slower but still safe when the network is struggling or under attack. Blue work is what makes it measurable, letting nodes compare total work across a graph instead of a single chain.
GHOSTDAG does not make every block equal. It gives nodes a rule for ordering the graph.
Two more parts keep that graph small enough to check. Storage-mass pricing charges a transaction for the UTXO footprint it leaves behind for good, on top of its raw size. Pruning and bounded anticones cap what a node has to hold and how far the parallel edges can spread, which is what lets an ordinary machine keep up with a fast proof-of-work DAG.
Hashdag's writing points that speed at shared commitments rather than at price charts. The simplest version is Kickstarter-like funding, where money moves only if enough people commit and a public rule holds the threshold instead of a company that can wave it away.
Any fast block-producing system can get a transaction into a block quickly. That's inclusion. It says nothing about confirmation: how fast the network gives strong confidence the transaction won't reverse. Mixing the two up is the most common way "fast blockchain" claims mislead.
Proof of Stake usually gets fast finality by quickly counting votes from staked validators, which ties confirmation speed to how well those validators coordinate and how the stake is spread. On Solana a stake-weighted schedule picks which validator leads each slot and puts the transactions in order. Push decentralization further and that job gets harder, which often drives the system toward committees or other sampling shortcuts.
Proof of Work samples security a different way. A block is proof the finder beat the rest of the network's hash power for that round, so the protocol never has to poll most of the miners before it confirms anything. Block speed and miner headcount come apart.
The two models also split on cost. Proof of Work brings outside cost into consensus, in power bills and hardware. Proof of Stake backs the ledger with token ownership inside the system instead. That gap changes who can censor, how new coins are issued, and what an attack costs.
"10 BPS" by itself is not the claim. K, finality depth, and merge depth all scale with the block rate, and they are what keep high block production from turning into runaway parallelism.
Kaspa's app-layer path runs through Toccata, spend rules, asset rules, Silverscript, ZK proof checks, sequencing commitments, and vProgs groundwork. Toccata activated at DAA score 474,165,565, the score set by Rusty Kaspa v2.0.0, roughly June 30, 2026 at 16:15 UTC. Michael Sutton's April 2026 outlook explains why the original May 5 target moved. The covenant rules are live on L1; wallet support, tooling, and app usage are the next evidence to watch.
The mechanism is that money can carry a rule forward. Transaction introspection, covenant IDs, sequencing commitments, and proof checks do that work in consensus itself, so a covenant-style app is a controlled UTXO state machine rather than a global account VM. Apps prove the richer logic around it. vProgs are a Kaspa-native verifiable-program design with their own proving and settlement assumptions, so calling them rollups brings in the wrong ones.
Builders get three things today and should say which one the product needs.
Payments, receipts, and node reads run on the live network now.
Vault, escrow, and asset rules map onto limited outputs now that Toccata is activated. Wallets and dev tools have not caught up, so the first builds bring their own.
Based apps take their order from Kaspa and prove results back to it, with no sequencer of their own. That is roadmap work, and "live" waits on wallet, tooling, and transaction evidence.
Naming a category is not shipping one. On Ethereum, "lending" means audited contracts, oracle feeds, and liquidation logic that have taken years of attack, so a Kaspa version needs its own custody model, oracle terms, liquidity, and proof that what it claims is really accepted on-chain. Kaspa brings no built-in liquidity or distribution either; a founder has to bring that plan. "Kaspa is fast" tells a reader nothing. "The vault needs a spend cap the protocol enforces" tells them why the network was picked.
The work sorts by how far along it is. Toccata's covenant rules are live on L1, though Kaspa-native DeFi built on top of them is still roadmap. DAGKnight is a research and upgrade track, vProgs are roadmap design, and RTD-derived oracle, TangVM, and attestation flows are app-level builds over basic Kaspa parts. Coordination markets stay a research direction until shipped products and repeat use exist.
Design is not adoption. On the network side, the case weakens if node costs rise too far at higher BPS, or if falling emissions aren't made up by real fee demand. On the app side, it weakens if Toccata's live rules fail to draw apps people use, or if bridge and oracle systems add trusted choke points that undercut the whole point. Rivals can also win users just by being good enough for the job, which is why the design claim only counts once it turns into wallets, exchanges, apps, and liquidity people use.
Checklist
Check price last.
Two answers do most of the sorting. If removing the token stops the network, it is a necessary native asset; if an existing coin or an ordinary database could run the same product, the token is decoration. And market cap is the number most often misread: price multiplied by circulating supply measures neither revenue nor capital invested nor how much could be sold without crashing the price. A market cap that looks small next to a much larger FDV is a dilution warning.
KAS clears the token-necessity question more easily than most app tokens do: it is the asset a PoW UTXO blockDAG network tracks, charges fees in, and pays out as the mining reward. Liquidity, security budget, mining concentration, node requirements, and app-layer execution are all still open, and the questions above are where a reader works through them. But remove KAS and there is no network left underneath it, and the fair launch leaves no insider allocation to discount for.