Venice AI Raises $65M at $1B Valuation as Private AI Moves Into the Mainstream

Venice AI has raised a $65 million Series A round led by Dragonfly Capital, giving the privacy-focused AI company a $1 billion valuation roughly two years after its public launch.

The Las Vegas-based company, founded in 2024 by Erik Voorhees, offers access to more than 200 AI models across text, image, video and audio. The round also included backing from Coinbase Ventures, North Island Ventures and others, and marks Venice’s first outside capital raise.

The funding comes as AI tools become more deeply embedded in how people search, write, code, create images, build agents and manage digital work. That shift has also raised a more uncomfortable question: who controls the data produced through everyday interactions with AI systems?

A Privacy-First Bet on AI Usage

Venice’s central argument is that AI will become too important to be treated like a conventional cloud software product. As people increasingly rely on AI systems for personal, professional and creative work, prompts can reveal sensitive information about a user’s business plans, health concerns, politics, finances, relationships or intellectual property.

Rather than asking users to trust that their data will be handled responsibly, Venice says it is designed around not possessing that data in the first place. The company says conversations remain on the user’s device, requests are relayed without being stored, and privacy protections vary depending on the model and mode being used. Its website describes a privacy architecture that includes anonymous access, private zero-data-retention inference, Trusted Execution Environment processing and end-to-end encrypted inference for Pro users.

That technical framing is what separates Venice from many AI interfaces that focus primarily on model quality, productivity or enterprise workflow integration. Venice is positioning privacy and user sovereignty as core infrastructure, not as a secondary feature.

More Than a Chatbot

Although Venice is often described as a private AI assistant, the company’s product has expanded well beyond basic chat. Its website lists support for text generation, image generation and editing, video creation, audio and music generation, coding, search and agent workflows.

The developer side is also becoming more important. Venice offers an OpenAI-compatible API that gives developers access to chat, image, audio, video and embedding models through a single API key. Its documentation describes support for more than 250 models and endpoints for tools such as web search, web scraping, file inputs and crypto RPC.

That makes the company’s opportunity broader than consumer AI. Venice is also competing for developers building agents, private coding tools, multimodal apps and AI products that need access to multiple models without routing all user activity through a single dominant provider.

The Model Router Thesis

One of the more interesting parts of Venice’s strategy is that it is not trying to build one foundation model and force users into that ecosystem. Instead, it gives users access to a wide range of frontier and open-source models, including models from OpenAI, Anthropic, Google, Mistral, Meta, Qwen, DeepSeek, xAI, Kimi, Black Forest Labs, Runway, ElevenLabs and others.

This model-router approach reflects where much of the AI market appears to be heading. Users increasingly want the best model for a specific task, not necessarily one default assistant for everything. A coding workflow may require a different model than image generation, long-context research, audio production or agent orchestration.

Venice’s bet is that privacy can become part of that model-selection layer. Users may not only choose based on speed, cost and capability, but also based on how much data exposure they are willing to accept for a given task.

Strong Growth Ahead of the Round

According to the company, Venice has grown to more than $70 million in annualized revenue, more than 3.4 million users and 85 billion tokens consumed per day since launching in 2024.

Those numbers help explain the size and valuation of the Series A. Venice is not raising on a purely speculative privacy thesis. It is raising after showing that a meaningful number of users are willing to use, and in many cases pay for, an AI product that treats privacy as a primary design principle.

The new capital will be used to scale Venice’s consumer app and API globally. That likely means more infrastructure, more model access, higher reliability, larger developer adoption and continued investment in privacy-preserving inference.

A Timely Debate Over AI, Surveillance and Control

The raise also lands at a moment when the relationship between AI and surveillance is becoming harder to ignore. AI systems are no longer just tools that answer isolated questions. They are becoming interfaces for work, search, creativity, software development and personal decision-making.

That creates a new kind of data trail. Prompts and AI interactions can be more revealing than search queries because they often include context, intent, drafts, files, business logic and private reasoning. For companies like Venice, this is the opening: if AI becomes a primary gateway to the digital world, privacy around AI interactions becomes a foundational issue.

At the same time, Venice’s unrestricted approach will continue to invite scrutiny. The company’s position on free expression and uncensored model access is part of its appeal to users who dislike centralized AI controls, but it also places Venice in the middle of a broader debate about safety, moderation and the responsibilities of AI platforms.