Sui reported that a live, high-stress demonstration at its Basecamp event in Singapore on October 7, 2026 reached 40,614,180 transactions per second through Sui tunnels and settled to Sui mainnet. The issuer said the test took place on the Summit Stage during its annual conference, which was held alongside Token2049 Singapore. The announced result is a performance claim from Sui, not an independently published audit conclusion.
The distinction between tunnels and ordinary standalone mainnet activity is central to reading the number. Sui said the reported throughput was measured inside Sui tunnels, where activity occurs offchain before a tunnel is closed and settles to mainnet. Readers therefore should not interpret the 40,614,180 TPS figure as a count of separate conventional mainnet transactions being processed individually at that rate.
According to Sui, more than 10,000 tunnels were opened during the demonstration in categories including chat, games and payments. The issuer described those tunnels as gasless and offchain during the test, while stating that closed tunnels are mutually cosigned and independently verifiable onchain. Sui also said its tunnel design is programmable, rather than being limited to payment use cases.
What the reported throughput measures
Sui described tunnels as comparable in one respect to Lightning Network offchain payment channels: they settle to the mainnet when they close. In Sui’s account, each closed tunnel is mutually cosigned and can be independently verified onchain. That architecture is why the demonstration can combine a very high reported rate of tunnel activity with settlement on Sui mainnet, rather than requiring every intermediate interaction to be handled as a separate mainnet transaction.
The issuer said a tunnel needs two transactions: one to open and one to close. Sui presented this model as a way to keep intermediate costs close to zero while allowing privacy and offline use. Those are issuer statements about the intended tunnel design and test context, not proof that every possible application or user will experience the same costs, privacy properties or operating conditions.
Sui said the Basecamp figure exceeded its previous reported record of 6,086,766 TPS, set on July 4, 2026. It characterized the newer result as more than six times the July figure and said both measurements were taken inside Sui tunnels. This comparison is useful because it uses the same announced measurement context, but it does not establish a general comparison with networks that measure throughput through different transaction types or execution environments.
The company said the event was designed to test resilience and speed under the extreme capacity demands it associates with AI agents and machine commerce. It also said participants and spectators could use an event-specific explorer while users and AI agents opened tunnels. The demonstration shows the type of activity Sui chose to test; the announcement does not independently establish broader demand, commercial adoption or performance across all blockchain workloads.
What remains to be independently checked
CertiK had not published an audit report when Sui made the announcement. Sui said CertiK, serving as an independent auditor, would examine associated data including proofs, execution logs, journals and supporting evidence, then publish a report in the coming days. Until that examination is completed and the report is available, the throughput result should be described as Sui-reported and audit-pending.
The pending review matters because a public performance claim involves more than a headline TPS number. The announced audit scope includes data and records that may help assess how the demonstration was conducted and what the reported result represents. The supplied information does not provide CertiK’s findings, validation methodology, conclusions on resilience, or any confirmation that the result applies outside the recorded test.
For developers, users and market observers, the immediate takeaway is narrow but concrete. Sui has announced a live tunnel-based test that it says reached 40,614,180 TPS and settled to its mainnet, with more than 10,000 tunnels opened during the event. The next material verification point is CertiK’s promised report, rather than an assumption that the announcement alone confirms the result or its wider implications.
- Treat 40,614,180 TPS as Sui’s reported tunnel-based test result, not as a direct measure of ordinary standalone mainnet transactions.
- Note that the event occurred in Singapore on October 7, 2026, during Sui Basecamp.
- Compare the result only with care: Sui said its prior July 4 record was 6,086,766 TPS, and both figures were measured inside tunnels.
- Watch for CertiK’s report, because its announced examination of proofs, execution logs, journals and supporting evidence was still pending.
| 項目 | Source-supported fact | 重要な理由 |
|---|---|---|
| Reported throughput | Sui reported 40,614,180 TPS in a live, high-stress test using Sui tunnels. | The figure describes activity measured inside tunnels. |
| Settlement model | Sui said tunnels settle to mainnet when closed and that closed tunnels are mutually cosigned and independently verifiable onchain. | Intermediate tunnel activity and mainnet settlement are not the same measurement category. |
| Prior reported record | Sui said its July 4, 2026 tunnel-based record was 6,086,766 TPS. | Sui characterized the Basecamp figure as more than six times higher. |
| Verification status | Sui said CertiK would examine test data and publish a report in the coming days. | An independent audit report had not yet been published at announcement time. |
この記事は情報提供のみを目的としており、投資助言ではありません。
admin_wcContributor · World-CryptMore from
