The raise is the company's first external capital since its founding in May 2024. By the time it closed, Venice was already running above a $70 million annualized revenue rate and had reached profitability, a profile that inverts the typical pre-revenue model that has defined most crypto-adjacent VC rounds. The company says it will use the funds to purchase GPUs and build own data centers, a move intended to improve gross margins over the longer term.

---

A Familiar Founder, A New Target

Venice AI was founded by Erik Voorhees, whose track record in crypto includes Satoshi Dice, one of Bitcoin's earliest gambling applications launched in 2012, and ShapeShift, the crypto exchange that pioneered non-custodial crypto swaps before regulators compelled it to adopt identity verification in 2018. Voorhees has applied the same data-minimization philosophy to AI. Venice runs inference without storing user prompts or outputs on centralized servers, and offers more than 200 models across text, image, video, and audio, including variants with fewer content restrictions than mainstream platforms.

"I think it's actually quite dangerous from a safety perspective, for the world to enter this next phase and have everyone be constantly watched," Voorhees said on the Unchained Crypto podcast. "To me that is actually much more dangerous than any particular person asking a controversial question, or something that might be considered bad." He has also flagged a more specific risk: that large AI companies could face pressure to modify model behavior to satisfy the Chinese government in order to access the market in that country, a concern he described as a reason to prefer infrastructure that is architecturally incapable of that kind of manipulation.

The platform serves 3 million active users and handles an average of 1.7 million API calls per day. Venice offers an OpenAI-compatible API, which means developers already building with other major providers can switch with minimal code changes.

---

The Token Layer

Venice has a native utility token, VVV, launched in January 2025 as an ERC-20 on Base, Coinbase's Ethereum layer-2 network. The token currently trades near $13, down roughly 39% from its all-time high of $21.32 reached on June 3, 2026. Market cap sits in the $584 million to $623 million range, placing VVV around rank 73 on CoinMarketCap as of this writing.

In 2026, Venice introduced a second token called DIEM. Users stake VVV to mint DIEM, which functions as a daily compute credit worth roughly $1 of API access, according to the company. The mechanics allow developers and AI agents to pay for inference through staking rather than per-request billing. Approximately 33.7 million VVV tokens, about 42% of the original 100 million supply, have been removed from circulation through burns and other mechanisms as of March 2026 (according to third-party on-chain trackers; Venice has not independently confirmed these figures). The company also cut its annual emission rate from 8 million to 6 million VVV per year in February 2026.

---

Why This Matters Outside the US

The fundraise lands at a moment when regulatory pressure on AI platforms is accelerating across several high-growth markets, and Venice's architecture positions it differently than compliant centralized alternatives.

In India, amended IT Rules that took effect in February 2026 impose a two-hour window for platforms to remove flagged deepfake content, require mandatory labeling of AI-generated material, and carry potential criminal liability for non-compliance. Centralized AI providers are building local compliance infrastructure to meet these requirements. Venice, which stores no content by design, sits in a structurally different compliance position, though Indian regulators have not yet addressed platforms with this architecture directly. India also added 24 new internet restrictions in 2025 and maintains a 30% tax on crypto assets, which creates friction for VVV token access at the retail level. For context, AI adoption across the Global South stood at 15.4% as of Q1 2026, compared with 27.5% in the Global North, a gap that frames the scale of the opportunity Venice is positioning toward.

In Nigeria, roughly 30 million people use crypto, most of them through peer-to-peer channels after the 2021 banking ban on crypto facilitation pushed activity off formal rails. That population is already comfortable with non-custodial, DeFi-native access layers, and Venice's staking model for compute access maps onto that behavior more naturally than a subscription billed to a card. AI adoption in Nigeria and Ghana sits at 10.1% and in Kenya at 8.7% as of Q1 2026, well below the global average, but the trajectory is upward and the demand for tools that do not route data through US or EU servers is real. South Africa presents a distinct profile: identified as Africa's AI adoption leader as of May 2026, it has a developed crypto regulatory framework under FSCA licensing since 2023 and a permissive AI policy environment, making it the most likely regional beachhead for any Web3-native AI platform expanding across the continent. Venice's 200-plus model library could in theory include open-source multilingual models relevant to underserved language markets across Africa, but the company has made no official announcement of Africa-specific or local-language model support, a gap that currently limits Venice's practical relevance in those markets to developer-tier users rather than general consumers.

---

What Comes Next

Dragonfly, which closed its own oversubscribed $650 million fourth fund in February 2026, exceeding its original $500 million target, and has backed projects including Polymarket, Ethena, and Avalanche, is making a clear bet that privacy infrastructure at the AI layer will carry the same strategic importance that privacy tooling carried in crypto's earlier cycles. Venice's GPU buildout will determine whether its margin profile can support the compute costs that come with scale. The more immediate test is whether the DIEM staking model holds developer interest as VVV price volatility continues. The token's 30-day swing from roughly $8.46 to $21 and back toward $13 is a reminder that the economics of token-gated compute are only as stable as the market underneath them.