Solana
KLEAR3 ResearchPublished 2026-08-25Updated
SOLSolana runs the most actively developed code of the ten profiles here: 1 215 commits from 30 distinct identities in 90 days. It also carries the strangest trap on this list, because the repository with 14,945 stars, the one every search returns, was archived in January 2025.
What does Solana actually do?
Solana runs programs on a chain built for throughput. It orders transactions with a clock that every validator can verify independently, which removes most of the back and forth other chains need before agreeing, and buys speed at the cost of much heavier hardware.
Where the chain that made programmable ledgers ordinary asks every node to be cheap enough that anyone can run one, Solana asks nodes to be fast. That single choice explains almost everything else about it, the low fees, the outage history, and the smaller number of machines that can keep up.
The trade is not hidden and it is not free. Fewer machines able to participate means fewer independent parties, and this page does not pretend that a faster chain is a better one. It is a different answer to the same question.
Source CoinGecko https://api.coingecko.com/api/v3/coins/solana, read 2026-08-25
Is the code actually open?
Yes, under Apache-2.0, with 179 releases. But the repository people find, solana-labs/solana, has been archived since 22 January 2025 and keeps all 14 945 stars. The maintained client is anza-xyz/agave, opened a year earlier and carrying a fraction of that visibility.
This is the fourth broken code trail on this list of ten, and the most instructive one. Nothing here redirects and nothing answers 404. The archived repository loads perfectly, shows real code, and is frozen, which is exactly what a reader checking whether a project is alive would take as proof that it is.
The stars are the reason it survives. Popularity is sticky and maintenance is not, so the dead repository keeps winning searches years after the work moved. Compare with an entry whose published address merely redirects: there, at least, the link lands on live code.
Source GitHub API https://api.github.com/repos/solana-labs/solana, read 2026-08-25
No audit report is published in the maintained repository, and none is listed. As elsewhere on this site, read that as an absence of reports rather than a verdict.
Source anza-xyz/security-audits https://api.github.com/repos/anza-xyz/security-audits/contents/solana, read 2026-08-26
Is anyone still working on it?
More than on anything else here. The maintained client took 1 215 commits from 30 distinct Git identities in the 90 days to 25 August 2026, the last on the day we measured. The next busiest profile on this list counted 367.
That figure comes from the pagination header rather than from counting objects, which matters: GitHub returns at most a hundred per page, so any count that lands on a round hundred is a ceiling. 1 215 is what the last page number says, and it is roughly three times the activity on the oldest chain here.
Volume is not quality, and we did not read the commits. What it does establish is that the client is under continuous work by a wide bench, which is precisely what the archived repository would have led you to doubt.
Source GitHub https://api.github.com/repos/anza-xyz/agave/commits?since=2026-05-27T00:00:00Z&per_page=1, read 2026-08-25
- The maintained client opens
- The old repository is archived
Where does it actually run?
On its own chain, which we checked responds: a public node returned slot 441 699 226 on 25 August 2026. It also hosts tokens issued elsewhere, so an address here can hold assets whose real story belongs to another page.
That second point is easy to miss. A dollar token held on this chain is the same issuer’s claim as the one held on Ethereum, but a different contract on a different ledger. The ticker travels, the risk does not follow it identically.
Source api.mainnet-beta.solana.com https://api.mainnet-beta.solana.com — getSlot, read 2026-08-25
Network verified by reading its current block height on a public node.
What the token actually does
SOL pays transaction fees and is staked by validators to take part in ordering blocks. There is no cap: 583 million circulate out of 633 million issued, and issuance continues on a schedule we did not examine or attempt to describe.
583,375,799 circulating against 632,859,657 issued is a narrow gap by the standards of this list, which is worth noting because the wide gaps get all the attention. Most of what exists is in circulation, and the open question here is the rate of new issuance rather than a locked reserve.
The same absence of a ceiling appears on another chain of similar age, and it means the same thing in both places: supply is governed by rules that can be changed rather than by a number nobody can move.
- What the token is forDirectly verified
- How many existThird-party supported
- How new ones appearNot verified
- Who can change the rulesNot verified
KLEAR3 rating
- Open code5
public repository · Apache-2.0 licence · 179 published releases · 1215 commits in 90 days
- Audits5
9 firms on the core · overlapping reviews · 3 audited scopes · report opened: 0 critical, 0 high, 0 medium · no critical or high finding · code reviewed in the open: 1215 commits in 90 days, 2 years in production
What holds it back audit stops at 2023-08-24: the code has moved since
- Business model5
business model published by the protocol · paid by le signataire de la transaction · collected by le producteur de bloc · la moitié des frais de base est détruite, le reste et les frais de priorité vont au producteur
- Governance4
decided by simd-process, with no vote · 123 improvement proposals published
- Token alignment1
supply measured, with no cap: nothing bounds issuance · circulating share known
What holds it back emission schedule unknown
- Maturity3
2 years of public code · 1 network verified by us
We rate the protocol: how it is built, how much of it can be checked, how its token lines up. Never an investment. A high score is not advice to buy, and a low one is not advice to sell.
What we could not establish
Governance, audits, the issuance schedule, and how validator participation is distributed in practice. We read the repositories and the chain, not the arrangements around them, and we do not describe what we have not opened. The distribution question is the one that would matter most.
The distribution question is the one that would matter most here. A chain that asks a lot of its machines narrows who can run one, and that narrowing is the whole cost of its speed. Measuring it properly needs a different campaign than this one.
Frequently asked questions
Which Solana repository should I actually read?
anza-xyz/agave, opened on 30 January 2024. The one search engines return, solana-labs/solana, was archived on 22 January 2025 and no longer receives work, despite holding 14 945 stars and most of the inbound links.
An archived repository is not a deleted one, and that is what makes this trap effective. It serves a complete, plausible, well-starred project page. Nothing on it looks stale unless you check the date of the last push.
The practical rule this suggests is worth generalising: stars measure attention over a project’s whole life, never its current state. The date of the last commit is the only field on a repository page that answers the question people ask stars to answer, and it takes one glance to read. Stars keep growing long after a project has stopped.
Why does Solana keep having outages?
We did not examine that history and we do not list incidents here. What this page can say is structural: a chain that pushes throughput asks far more of each machine, and the same design that produces low fees also produces less margin when something goes wrong.
Treating outages as a scandal or as a non-event both miss the point. They are the visible price of an explicit engineering trade, and the honest way to weigh them is against what the trade buys.
For a reader deciding what to hold, the question is not whether a chain has ever stopped but what happens to their position when it does. That answer depends on the application, not on the chain, and this page does not cover applications, only the chain beneath them.
Is Solana faster than Ethereum?
On raw throughput, yes, by a wide margin. That comparison is also the least interesting one, because the two chains have made opposite choices about how cheap a participating machine should be, and speed is the visible half of that decision.
The useful question underneath is how many independent parties can afford to take part. A ledger is only as decentralised as the cost of running a node, and that cost is a design input, not an accident.
Both chains run programs, both host tokens issued by others, and both are actively developed. Choosing between them on speed alone answers a question almost nobody actually has. What a holder wants to know is what would have to fail for their position to become unreachable, and that answer lives in the application layer rather than in a throughput figure.
Next: our Hyperliquid profile shows the same speed trade taken further, with no protocol code published at all, and our BNB profile shows a chain that bought speed by reducing the number of validators instead.
Ask an assistant to summarise it and check what it claims.