Antelope Chain Status

What each Antelope mainnet is running today, measured from its own public API nodes, and what that means for the state-import migration path. This is a compatibility reference, not a roadmap of who is migrating.

Surveyed 2026-09-07 with get_info and get_activated_protocol_features against several public nodes per chain (57 nodes answered across eight chains). A single node reports itself, not the network, so every row is a multi-node reading.

Status at a glance

Chain	Nodes	Node software	Finality (head − LIB)	Features	PulseVM import path
XPR Network	11	Leap 5.0.3 (7), mixed older	DPoS, ~330	19	Demonstrated
WAX	13	Leap 5.0.x (10), Spring 1.3.x (3)	DPoS, ~135	19	Same path as XPR
Libre	5	Leap 5.0.x (4), one 3.2.3	DPoS, ~330	14	Same path as XPR
EOS / Vaulta	7	Spring 1.0.5 → 1.2.2	Savanna, 2	23	Same path, roadmap items
Telos	8	Spring 1.0.3 → 1.2.2	Savanna, 2	23	Same path, roadmap items
Ultra	4	own fork, v6.2.2	Savanna, 2	26	Not assessed
FIO	9	own lineage, v3.5.1	DPoS, ~330	0	Not assessed
UX Network	0	not reached	—	—	Not reached

What the labels mean

• Demonstrated — a real snapshot of that chain's state has been imported onto a PulseVM node and transacted against with the chain's existing keys. The 1:1 demo network is a byte-exact import of XPR testnet state.
• Same path as XPR — the chain writes the same snapshot format (chainstate version 6, from any Leap 5.0.x node) and activates the same, or a subset of the same, protocol features as XPR. The demonstrated import applies without new reader or host-function work. Nobody has yet run the ceremony against that chain's state.
• Same path, roadmap items — the contract state imports the same way, but the source runs Spring with Savanna, which adds three tracked items: reading chainstate snapshot version 8, serving the bls_* host functions of BLS_PRIMITIVES2, and accepting the system contract's set_finalizers / set_proposed_producers_ex calls as recorded no-ops, since they have no consensus meaning on Snowman. Savanna chains already have deferred transactions disabled, which removes an import step rather than adding one.
• Not assessed — the chain runs a fork that diverges from the Leap/Spring line in ways this survey cannot classify from the outside.

Per-chain notes

XPR Network. Seven of eleven nodes run Leap 5.0.3; the rest report 5.0.0, 5.0.2, 3.1.2 and one Spring-based build serving without Savanna. The 19 activated features are the reference set PulseVM's import carries.

WAX. All nodes run WAX's wax01 builds. Ten are Leap-based (5.0.3 on six, plus 5.0.0, 5.0.1 and 4.0.5) and three are Spring-based 1.3.x, with Savanna not activated. The feature set is identical to XPR's, and any Leap 5.0.x node can write a version-6 snapshot today.

Libre. Leap 5.0.x on four of five nodes. Fourteen activated features, a subset of XPR's. Libre's custom system contracts would need review before a rehearsal.

EOS / Vaulta and Telos. Every node runs Spring, from 1.0.5 (EOS) or 1.0.3 (Telos) up to 1.2.2, with a few main-branch builds reporting 1.3.0 that correspond to no tagged release. Savanna is active on both, so the last-irreversible block trails head by two. Both activate the same 23 features: XPR's 19 plus SAVANNA, BLS_PRIMITIVES2 and the two DISABLE_DEFERRED_TRXS stages. No Leap nodes remain, so a snapshot from these chains is version 8.

Ultra. Ultra's own fork at v6.2.2-3.0.0 with Savanna active and 26 features, the EOS set plus Ultra-specific ones. The Savanna items apply, and the fork-specific features need their own assessment.

FIO. FIO's own lineage at v3.5.1, predating the protocol-feature framework, reporting zero activated features. A diverged fork that would need its own snapshot reader and feature audit.

UX Network. None of fourteen candidate public endpoints answered in this survey.

Why the source's consensus does not matter

The migration imports chain state: accounts, permissions, contract code, tables, balances. Consensus state is discarded, because finality on PulseVM comes from the Avalanche Snowman engine, where a block is final on acceptance and last-irreversible always equals head. Whether a chain has activated Savanna changes the snapshot format and a handful of host functions, not the outcome: a chain arrives with finality of about a second and no reversible window either way. See the FAQ on the migration guide.

Reading the numbers

• Finality is head minus last-irreversible block as reported by the nodes. Around 330 blocks is classic DPoS finality, about 165 s at 0.5 s blocks; 2 blocks is Savanna.
• Features is the count from get_activated_protocol_features. Two of XPR's, CRYPTO_PRIMITIVES and GET_BLOCK_NUM, are activated on XPR and WAX but not yet served as host functions by PulseVM; an audit of every contract deployed on XPR testnet found none that import them, and serving them is tracked work.
• Node software mixes are normal on a live network. A chain can run Spring-based nodes without activating Savanna, as WAX does today.

Related

• Migrating an Antelope Chain to PulseVM — the path, the rehearsed ceremony, and the FAQ
• Antelope Compatibility — host-function and tooling surface
• The 1:1 Demo Network — the demonstrated import, live