What is Kaspa (KAS)?

Bitcoin's design has one structural cost that nothing about it can remove: only one block can win. If two miners find a block at the same moment, one is orphaned and the work that produced it is discarded. Keeping that waste small is why blocks are ten minutes apart, and that interval is why confirmations are slow.

Kaspa's answer is to stop discarding. GHOSTDAG arranges blocks as a directed acyclic graph rather than a chain: parallel blocks are all kept and consistently ordered instead of one being thrown away. Nothing is wasted, so blocks can come far faster. Kaspa runs at ten per second, which means confirmation in about a second while remaining proof of work throughout.

This is a genuine contribution to consensus design rather than a parameter tweak, and it is why Kaspa gets taken seriously by people who dismiss most Bitcoin alternatives.

The fair launch

Kaspa went live in November 2021 with no premine, no ICO, no venture allocation and no founder tokens. Every KAS was mined, by anyone who pointed hardware at it, from block one. Among assets launched after 2017 this is close to unique, and it means there is no unlock calendar, no investor cost basis near zero and no treasury overhang.

The 2026 upgrades

Crescendo, in May 2025, took block production to ten per second. Toccata, in late June 2026, added what Kaspa had conspicuously lacked: native smart contracts, KRC-20 tokens on the base layer, and zero-knowledge verification built into the protocol.

That turns Kaspa from a fast payments chain into something programmable, and it does so on a network whose issuance is almost finished — which is an unusual sequence. Most chains build the programmability first and hope the token economics follow.

• Payments and transfers, with roughly one-second confirmation at negligible cost.
• Mining rewards, still the only way new KAS enters circulation — and nearly exhausted.
• KRC-20 tokens and smart contracts on the base layer, live since the Toccata upgrade.
• Fees, which will have to carry miner revenue once issuance ends.

Kaspa's audience actually mines, which is unusual and shows up in what people search for: calculators, rigs, profitability. The network went from GPU-mineable to ASIC-dominated as it grew, the same trajectory Bitcoin and Litecoin followed, and the hardware question is the live one for anyone joining now.

28.7 billion KAS is the hard cap, and 27.37 billion — 95.39% — is already mined. New issuance is approaching zero by late 2026.

The chromatic emission schedule

Rather than halving the reward every four years in one step, Kaspa multiplies it by 2^(-1/12) every month. Over twelve months that compounds to exactly a halving, delivered as a smooth decline instead of a cliff.

It is a better design than Bitcoin's and it deserves saying why: a step halving cuts miner revenue in half overnight, forcing a sudden consolidation among operators whose economics no longer work. A smooth decline lets the industry adjust continuously. Kaspa took Bitcoin's monetary policy and removed its one obviously avoidable shock.

The security budget question, arriving early

The flip side is that Kaspa reaches the end of issuance far sooner than Bitcoin does. Once the subsidy is effectively zero, transaction fees have to pay for all security — and Kaspa's fees are deliberately negligible, on a network whose whole pitch is cheap fast payments.

Bitcoin faces this question in decades and has not answered it. Kaspa faces it within a few years. The smart contracts and KRC-20 tokens added in 2026 are, read economically, an attempt to generate fee demand before the subsidy runs out. Whether they generate enough is the central question about this asset and it will be answered soon rather than eventually.

Kaspa cannot be staked. It is proof of work with no bonding mechanism and no protocol yield.

Mining is the only way the protocol pays anyone, and with issuance nearly exhausted that payment is shrinking toward fee revenue alone. Products advertising KAS staking are lending arrangements or wrapped positions on other chains, with counterparties the network itself does not have.

The security budget runs out soon

At 95.39% mined with issuance approaching zero by late 2026, Kaspa is about to find out whether a chain with deliberately negligible fees can pay for proof-of-work security. This is the same problem Bitcoin has, arriving roughly two decades earlier, on a network with a fraction of the fee revenue. It is the most important thing on this page.

Hashrate is small in dollar terms

Kaspa's security budget is far smaller than Bitcoin's, and as the subsidy declines it shrinks further. A proof-of-work chain with falling miner revenue becomes cheaper to attack, and the Monero episode of August 2025 showed what a funded actor can do to a network whose hashrate is affordable.

New programmability is new attack surface

Smart contracts, KRC-20 tokens and ZK verification all arrived in one upgrade in June 2026. Kaspa's safety record to that point rested substantially on doing very little. Every chain that has added contracts has found bugs afterwards, and Litecoin's April 2026 MWEB exploit is the cautionary example from a directly comparable project.

No treasury and no funding

The fair launch that gives Kaspa its clean distribution also means no foundation budget, no grants programme and no funded development organisation. Work depends on contributors and donations, which has produced impressive results and is not a durable model if key people leave.

Competing against an entrenched answer

Kaspa is technically better than Bitcoin at fast cheap payments, and that market went to stablecoins on cheap chains rather than to a volatile proof-of-work asset. Being the best implementation of a category the market moved past is Litecoin's story too.