Base’s Beryl upgrade: B20 standard could change onchain assets
Base’s Beryl upgrade is scheduled for mainnet on June 25. The headline piece is B20, a token standard for stablecoins and real world assets on Base. My take: this is less about a shiny token format and more about making Base look usable to issuers that cannot wing compliance.
Base says it wants more institutional users and deeper liquidity. Maybe. I’ll be honest: that line is easy to say and hard to prove. The upgrade is already live on Base Sepolia, and it also shortens the usual Base-to-Ethereum withdrawal delay from seven days to five. It adds Reth V2, which should make nodes less expensive to run. Beryl follows Azul, which shipped four weeks earlier, moved Base to the Base Stack, and added a multiproof withdrawal system.
B20 is a native token standard. Its code sits inside Base node software instead of existing only as regular smart contracts. That is the important bit.
Base says B20 follows ERC-20, so wallets, exchanges, explorers, indexers, and onchain protocols should already know how to handle it. Why does this matter? Because the fastest way to kill a new token standard is to make every app write special support for it. The issuer toolkit includes pausing, supply caps, transfer limits, signature approvals, transfer memos, and the ability to freeze or burn assets held by blocked addresses. Base and Spearbit audited the toolkit. Most guides frame compatibility as the whole story. That’s only half right. The real signal is that Base is packaging controls for issuers that need compliance before they move assets onchain.
B20 launches in two versions: Asset and Stablecoin.
Base says the Asset version is for general token issuance. It supports configurable decimals, issuer metadata, event announcements, and rebasing. The Stablecoin version is for fiat-backed assets, with six decimal precision and an issuer-selected currency code. That gives banks and payment companies a more direct way to issue regulated stablecoins on Base, assuming they want to use it. My read: the assumption is doing a lot of work there.
Tokens issued through B20 should behave like normal ERC-20 assets, so apps should not need separate integrations.
Base is betting that plain compatibility will help adoption. It also plans to let users pay transaction fees with B20 tokens instead of ETH. Planned features include virtual deposit addresses. Direct node access to balances, transfer history, and token metadata is also on the list. Is this just convenience polish? Not really. In practice, that could make Base less awkward for everyday users and cheaper for traders who move assets often.
Base expects native execution to cut transfer costs by about 50% over time and double transfer throughput.
For traders and dApp users, that means faster and cheaper transfers. Thin margins notice. The withdrawal delay cut, from seven days to five for the more common single proof process, also frees up capital sooner. Counter to the usual advice, the flashier one-day withdrawal path is not necessarily the one to watch. Azul’s one-day path still exists, but Base says it is rarely used because generating zero-knowledge proofs is expensive.
Reth V2, part of Beryl, should reduce disk usage for full, minimal, and archive nodes. It should also speed up state root calculation.
Base says these changes should let the network raise block gas targets and limits without putting too much load on the sequencer or node infrastructure. This is the quieter part of the upgrade, but I would not treat it as secondary. If activity picks up after June 25, node cost and state calculation speed become less abstract very quickly. L2s do not get mass usage on vibes. They need capacity.
Regular users do not need to do anything. Node operators need to update their software before Beryl activates on June 25. Skip this step, and the upgrade is no longer theoretical.
Base says its next upgrade, Cobalt, is targeted for September. It is expected to add native account abstraction, expand B20, and merge the network’s consensus and execution clients into one node binary. Yes, this sounds like the usual roadmap pile-up. Still, if Base can ship Cobalt without making the network harder to operate, it gets a stronger case against other L2s.
What this means
Beryl, and especially B20, shows Base making a more obvious push toward institutions and enterprise asset issuance. I would put the emphasis on “issuance,” not just “institutions.” Those are different bets.
The built-in controls give it away. Pausing, freezing, supply caps, blocked-address handling, and audited issuer tooling are not mainly for meme coins or weekend experiments. They are for issuers with lawyers, risk teams, compliance checklists, and people who ask what happens when an address must be blocked. If financial institutions use B20 for stablecoins or tokenized assets, Base could see more total value locked and more demand for apps built on top of it. For crypto investors, the practical move is to watch projects that adopt B20 early. Some will probably sell the institutional angle too hard. A few may actually get traction. Lower costs and better throughput could also make Base more useful for DeFi, which may increase transaction volume and indirectly affect ETH through Ethereum’s fee burn.
What to watch next: June 25, when Beryl is scheduled to activate.
The first thing to track is B20 usage. New stablecoin launches matter. So do RWA announcements, issuer names, and actual onchain volume. Press language is the weak signal here. TVL on Base in the weeks after June 25 will give a cleaner read on whether the market cares. September is the next date on the calendar because Cobalt is supposed to bring native account abstraction and more B20 functionality. That could make Base easier to use and more competitive with other L2s. Traders should also watch Base transfer costs. If the promised 50% reduction shows up in real activity, arbitrage and higher frequency strategies become more attractive.