Sparkz is tokenless by default. A Capsule accumulates backing, receipts, and momentum long before anyone asks whether it should have a coin. But when a Capsule does earn that moment - real community, real funds, real status - the graduation has to be clean: no migration, no fragmented liquidity, no starting over. That is exactly what Clanker is built for, which is why it is the token rail Sparkz graduates onto.
What Clanker is, and where it stands
Clanker is a Farcaster-native token launcher on Base. You can deploy a token by mentioning a bot, through its web UI, or programmatically through its TypeScript SDK - and get a live token with liquidity in under a minute.
A note on versions, because it matters for planning. As of this writing the current production release is Clanker v4 (the clanker-sdk is at v4.2.x on npm). v5 is not yet shipped - the team sent it to audit in late June 2026, and it will ship with the B20 standard as default. So the honest picture is: build on v4 today, plan the v5 upgrade for when the audit clears. v4 is fully backward-compatible, so this is an upgrade path, not a rewrite.
Why the mechanics fit "graduate later"
Three things about Clanker's v4 model make the graduation clean:
It is one programmatic call. The SDK exposes a deployTokenV4 method. Sparkz can call it from the backend on an explicit operator action - a "graduate" button - with the full config in one shot: fee tiers, vault/lock, reward recipients, metadata. No listing process, no manual liquidity dance.
Fees route into a split you already have. Clanker lets you set reward recipients with fixed splits. A graduating Capsule already has an Empire treasury - a 0xSplits contract on Base - so the token's trading fees can flow straight into the same split that the tokenless Capsule was already using. Contributors, reserve, and creator keep their shares whether or not a token exists. That is what makes "the coin is optional" true in the plumbing, not just the pitch.
Liquidity is locked, so backers are protected. Clanker's LP is deposited to a locker and cannot be withdrawn - a permanent-liquidity guarantee that removes the most common rug vector. For a platform whose whole thesis is "back the work, not a coin," launching onto a rail that structurally cannot rug your backers is the point.
Creator-first, by default
Clanker's default reward split favors the deployer, and with an interface partner in the mix the split becomes creator / partner / protocol. That lines up with the Sparkz creator-first stance (our 1/1/98 default) rather than fighting it. Every trade then does double duty: it pays the creator and their contributors, and it writes more data into the Capsule - who bought, when, at what price. The token becomes another stream into the moat, not a replacement for it.
What this means for Sparkz today
- The Empire (tokenless create2 + 0xSplits treasury) is the pre-token home. It exists
from day one and nothing is on-chain until the first interaction.
- The Clanker config seam is already in the Capsule's
economic_config, ready for the
graduation call.
- Ship on v4 now; adopt v5 (B20 default) when it clears audit.
A token should be the thing a spark earns the right to launch - and when it does, it should be a single call that preserves everything the Capsule already built. Clanker is the rail that makes that true.
Sources: the current clanker-sdk release and v4 documentation confirm v4.2.x as current and v5 as in-audit at time of writing. Fee percentages and protocol economics evolve - treat specific numbers as a snapshot, and the v4-to-v5 upgrade as the near-term watch item.