Soda Labs closes $3M seed round for private blockchain transactions

• Soda Labs raised $3 million from NextBlock in October 2026, entirely from the Luxembourg venture firm’s $40 million inaugural fund.
• Soda Bubble, a chain-agnostic coprocessor using garbled-circuit multiparty computation, now processes private workloads on Ethereum, Arbitrum, Base and Polygon.
• Soda Labs reports 500 confidential transactions per second sustained on Arbitrum at $0.14 per million transfers, with five- to tenfold improvement over prior testing.

• $3M Seed round from NextBlock, entirely funding the October 2026 close
• 500 tps Confidential transactions per second sustained on Arbitrum protocol
• $0.14 Cost per million transfers on Arbitrum deployment

Soda Labs disclosed a $3 million seed round from NextBlock, a Luxembourg-based venture firm, to accelerate the transition from pilot deployments to production-grade privacy infrastructure for banks and payment networks. The capital, first reported by Cryptopolitan, arrives as the firm prepares to migrate its core technology, Soda Bubble, a chain-agnostic coprocessor that processes encrypted transactions without exposing balances, amounts or counterparties, from research to revenue operations. Soda Labs co-founder and CEO Avishay Yanai said the round would allow the firm to “take it from pilots to production,” marking the first institutional validation of garbled-circuit multiparty computation (GC-MPC) as a viable scaling layer for confidential enterprise blockchain traffic.

NextBlock backs Soda Labs with full seed commitment from $40 million Luxembourg fund

NextBlock, which launched its first Luxembourg-regulated alternative investment fund in July 2025 with $40 million in initial commitments, directed all $3 million of Soda Labs’ seed round from that vehicle. The fund, managed by an authorised Luxembourg AIFM licensed by the CSSF, invests in early-stage European and American blockchain companies and seeks exposure to both traditional crypto assets and decentralized finance applications. NextBlock founder and general partner Pieter van Poecke cited Soda Labs’ existing paying customers, technical intellectual property and “strong commercial instincts” as the driver behind the all-in commitment, signaling confidence in both the technology and the team’s ability to convert pilots into contracts.

The fund holds plans to increase capital to $60 million, reflecting institutional appetite for regulated privacy infrastructure that can serve banks without exposing them to token volatility or permissionless chain risks.

Soda Bubble’s encrypted computation now active on four major EVM chains

Soda Bubble operates as a multiparty-computation layer that performs calculations directly on encrypted data, splitting information across independent validator nodes so no single operator can decrypt it.

The system achieves confidentiality, not anonymity, by allowing authorized parties such as banks or regulators to decrypt transaction details after settlement, while keeping amounts and counterparties hidden from the public blockchain record.

Soda Labs’ stack runs on standard cloud CPUs using conventional AES and SHA256 encryption, requiring no specialized hardware, and has been audited by Hacken and covered by three granted US patents.

The technology is now active on Ethereum, Arbitrum, Base and Polygon, with Solana integration planned for a future date. Soda Labs reported achieving 500 sustained confidential transactions per second on Arbitrum during July 2026 testing across 15,000 transfers, at a cost of $0.14 per million transfers, five to tenfold better performance than its prior testing cycle.

The firm committed to publishing updated benchmarks in the coming weeks, setting a specific date for verifying claims that have become routine assertions across privacy-focused builders.

This cost and throughput profile directly competes with Tether’s confidential USDT deployment on Zama, which uses fully homomorphic encryption, suggesting two distinct cryptographic approaches are emerging for institutional privacy.

Soda Labs expands beyond COTI ecosystem roots into multi-chain institutional deployment

Soda Labs’ garbled-circuit technology has been live since March 2025 on COTI’s mainnet, where it became the first recipient of COTI’s $25 million ecosystem fund in 2024. The move to Soda Bubble and deployment across four separate blockchains represents a deliberate pivot toward vendor neutrality, decoupling privacy services from any single protocol’s success. 21Shares named privacy as “the missing layer between blockchains and institutions” in September 2026, positioning confidentiality not as a niche feature but as essential infrastructure for regulated finance.

Competing approaches are crystallizing: EthSystems, launched in July 2026 by former Ethereum Foundation researchers, is pitching privacy infrastructure directly to banks. Charles Hoskinson has pitched Midnight’s Night Mode privacy layer, while Aave’s ecosystem is exploring how privacy could enable institutional lending at scale. The market is fragmenting, not consolidating, across encryption methodologies, a pattern that may force banks to support multiple privacy stacks simultaneously rather than betting on one winner.

The CCS read. We see the infrastructure play, not the token thesis. Soda Labs is raising as a service provider to institutions that want on-chain settlement with confidentiality, not as a protocol asking for fee-based governance or yield accrual to holders. The real winner here is NextBlock’s ability to back infrastructure before it becomes a competitive commodity, and the real risk is whether banks will actually deploy beyond pilots once regulatory clarity forces their hand.

Soda Labs has committed to publishing updated performance benchmarks within weeks, a specific deliverable against which institutional buyers can measure the technology’s readiness for production. The open question is whether the firm will disclose which banks or payment networks are now moving beyond pilot status into live transaction volume, and whether throughput and costs scale as confidential workload enters production rather than test environments.