Blockchain Upgrades Weekly: September’s Big Moves From Solana To 0G

by
Alisa Davidson

Revealed: September 30, 2025 at 10:25 am Up to date: September 30, 2025 at 10:25 am

by Ana

Edited and fact-checked:
September 30, 2025 at 10:25 am

To enhance your local-language expertise, typically we make use of an auto-translation plugin. Please observe auto-translation is probably not correct, so learn unique article for exact info.

September didn’t give us a lot time to breathe, did it? As a substitute of easing into autumn, the blockchain area hit us with a type of months the place practically each main community appeared to both launch one thing shiny or overhaul its insides. And whereas a few of these upgrades scream hype, others are the sort that quietly reshape how issues really work — the unglamorous plumbing that no one tweets about however everybody ultimately advantages from.

Let’s sit down and untangle what actually mattered.

Solana — Alpenglow improve

We’ve bought to start out with the apparent heavyweight this month. Solana’s Alpenglow is out, and it’s not as a lot of an improve as a full-on character shift. Proof of Historical past and Tower BFT — the very elements which have outlined Solana’s id thus far — are being phased out in favor of Votor and Rotor. Catchy names apart, what they promise is staggering: transaction finality in roughly 150 milliseconds.

Now, take into consideration what which means. For merchants working high-frequency bots, that’s basically close to parity with centralized exchanges. For gaming initiatives, it might imply actually fluid gameplay the place blockchain logic doesn’t really feel like a clunky add-on. However — and right here’s the skeptical voice within the room — do we actually want finality that quick? Most customers can’t even understand 12 seconds as a delay, not to mention a fraction of 1. So the query is: does Alpenglow flip Solana into a series for specialised, high-speed use circumstances, or does it merely gild the lily? Both method, it’s one of many boldest architectural overhauls of the 12 months, and that makes it inconceivable to disregard.

Arbitrum — Stylus on mainnet

Then we shift gears to Arbitrum. Stylus went stay on mainnet, and although the improve is much less flashy, it would find yourself extra impactful in the long term. Solidity has at all times been Ethereum’s velvet rope — you need in, you study it. However now? Rust, C, and C++ devs can deploy sensible contracts immediately on Arbitrum.

That is large. Consider all of the builders on the market in Web2 land who’ve been constructing in Rust or C++ for years — immediately, they will wade into Web3 with out studying a brand-new language. Will they really come? That’s the open query. Tradition shock is actual, and Solidity devs nonetheless dominate the panorama. However the mere chance of reducing that barrier means Arbitrum might turn into the place the place Web2 meets Web3 in apply. Arbitrum is mainly betting on increasing the expertise pool. And in an area that always suffers from echo-chamber considering, that could be the neatest wager.

Starknet — model 0.14.0

Starknet’s September transfer was quieter however sensible: six-second block occasions, a brand new mempool, and a distributed sequencer. These sound technical — and they’re — however in apply, it means fewer moments the place you’re ready, watching, and muttering “come on, affirm already.”

It additionally tackles one of many community’s lingering PR issues: being seen as clunky in comparison with different rollups. If they will clean out that person expertise, builders could lastly cease making excuses for Starknet’s “potential” and begin constructing with out caveats. Nonetheless, right here’s the lingering doubt: are six-second blocks sufficient to make customers care, when rivals like Solana are boasting sub-second finality? Starknet isn’t actually chasing pace data, however fairly concentrating on stability and credibility. Typically, regular enchancment is strictly what a rollup wants — even when it doesn’t seize the highlight.

Stellar — Protocol 23 (Whisk)

Stellar usually feels just like the quiet child within the nook: dependable, well mannered, however not often the middle of the dialog. After which Protocol 23 landed. Immediately, exchanges like Upbit have been pausing buying and selling throughout the rollout, which tells you right away — that is necessary.

Parallel transaction processing is the large deal right here. Stellar can now deal with extra without delay, smoothing the pathway for scale. Soroban sensible contract optimizations are the cherry on prime, making it simpler for builders to deploy apps with out worrying about hitting bottlenecks. What’s refreshing is that Stellar isn’t chasing traits like AI or gaming — it’s doubling down on funds, the factor it’s at all times been good at. The larger query: is “being good at funds” sufficient to make noise in 2025, when everybody else is pitching futuristic visions? Possibly not, however this improve ensures Stellar remains to be within the working.

Hedera — v0.65

Hedera had its personal flip with v0.65 on the finish of September. Digital Mega Map, refined price schedules, smoother node efficiency — in brief, the kind of modifications that enterprises clap politely for as a result of they imply much less friction behind the scenes.

Let’s be actual: nobody’s speeding to purchase HBAR on the again of a Digital Mega Map. However that’s Hedera’s model — it doesn’t cater to hype cycles, it caters to companies that need blockchain to work. If you consider it, somebody in an enterprise IT division might be relieved proper now as a result of Hedera shaved just a few milliseconds off some course of or diminished node crashes. Does that make headlines? No. Nevertheless it does matter for adoption within the slow-and-steady company lane.

0G (Zero Gravity) — Aristotle mainnet launch

Now for the flashy one. 0G launched its Aristotle mainnet mid-month, and also you couldn’t ask for a greater combo of buzzwords: AI, modular chains, and fast change listings. Storage, compute, inference — all spun off into their very own devoted chains. And the market liked it, pumping its cap previous $1.2 billion nearly in a single day.

Right here’s the place I put my skeptical hat on: is that this actual infrastructure or simply narrative chasing? Crypto has a monitor report of fusing itself to scorching tech traits — keep in mind the “metaverse” craze? However right here’s the counterpoint: AI really does want decentralized compute and storage in ways in which centralized incumbents can’t or received’t at all times present. So possibly 0G isn’t simply hype. Possibly it’s the canary within the coal mine for the place the subsequent wave of demand is coming from. Both method, Aristotle’s launch reveals simply how hungry the market is for the AI + crypto marriage.

Plasma — mainnet beta

Plasma went stay with a beta mainnet on September 25, conveniently timed with Korean Blockchain Week. And whereas most new L1 launches blur collectively lately, Plasma’s angle is evident: stablecoin funds.

Which may not sound horny, however give it some thought: stablecoins are already the lifeblood of crypto — trillions in annual switch quantity — and but most blockchains deal with them as simply one other token sort. Plasma is mainly saying: “Let’s construct an entire chain for this one function.” It’s targeted, nearly minimalist. Will it take off? That relies upon totally on whether or not customers care sufficient about saving fractions of a cent on funds, and whether or not retailers really undertake it. For now, it’s extra of a seed than a headline, however seeds can sprout quick on this area.

Boundless — ZK compute market on Base

And eventually, Boundless. This one didn’t make retail chatter, however from an infrastructure standpoint, it’s really fascinating. It’s a market for zero-knowledge computation, working on Base. Think about you’re a undertaking that wants heavy computation — as an alternative of burning cycles your self, you outsource it to provers who run it off-chain and hand you again validity proofs. Incentivized by Proof-of-Verifiable-Work, no much less.

It could really feel nerdy, however integrations with Wormhole, Lido, and others present it’s not simply concept. The query this raises is: how lengthy earlier than we take zk-based compute with no consideration, the identical method we now take L2 rollups with no consideration? Boundless could be a type of launches that doesn’t seem like a lot right this moment, however a 12 months from now we glance again and go, “Ah, that’s when the groundwork was laid.”

So, what’s the takeaway?

September was a buffet of upgrades: some glamorous, some quietly important, some speculative. And possibly that’s the healthiest signal but — there is no such thing as a single “path” for blockchain anymore. As a substitute, we’ve bought parallel bets being made on completely different futures. Will all of them work? Undoubtedly not. However the variety itself is proof that the ecosystem isn’t slowing down; it’s widening out.

Alisa, a devoted journalist on the MPost, focuses on cryptocurrency, zero-knowledge proofs, investments, and the expansive realm of Web3. With a eager eye for rising traits and applied sciences, she delivers complete protection to tell and interact readers within the ever-evolving panorama of digital finance.

Extra articles