$150M/year free speech Higgsfield
Jan-Erik Asplund
TL;DR: Venice is bringing the free-speech, anti-woke positioning of alternative social networks like Gab & Truth Social to AI, bundling private, uncensored chat, image & video generation. Sacra estimates Venice hit $150M in annualized revenue in September 2026, up from $10M at the end of 2025. For more, check out our full report and dataset on Venice.
We’ve covered how conflicts over free speech & content moderation create openings for new platforms, from Reddit’s clashes with its users (June 2023) to Truth Social & Gab building alternative social networks (March 2024) and xAI bringing the same free-speech positioning to AI with Grok (January 2025).
In AI video, we’ve interviewed Runway CEO Cristóbal Valenzuela on the emerging field of AI-generated video (January 2023), OpenArt CEO Coco Mao on building the TikTok for AI video (February 2025) and Higgsfield CEO Alex Mashrabov in on orchestrating AI video models together (January 2026).
Now we’re looking at Venice ($65M raised, Dragonfly), which has built a multimodal AI gateway around the core principle of private, uncensored creation of text, images, and videos.
Key points via Sacra AI:
- Amid the early 2024 backlash to perceived safetyism & bias across corporate AI models—with Gemini’s racially diverse Nazis as the poster child—Erik Voorhees, founder of “no-KYC” crypto exchange ShapeShift (2014), launched Venice AI (May 2024) as the right-libertarian ChatGPT alternative, making it easy to access uncensored variants of open-source models across text (Meta’s Llama 3) and image generation (Playground 2.5, Stable Diffusion). Rather than set up & run open-source AI models themselves on their own hardware, users can access them through Venice’s cloud service with zero data retention, paying in dollars or by connecting a crypto wallet with DIEM, its experimental token that provides users & AI agents $1/day recurring access to API credits, with VVV as the capital asset that earns yield, mints DIEM and accrues value by revenue-funded token buyback-and-burn.
- When xAI tightened safeguards on “free speech” Grok Imagine’s website & mobile app following backlash to deep faking (January 2026), its API remained more permissive, letting Venice package access to Grok’s image & video generation as Grok Imagine Private & driving 100% MoM growth in video generations in April & again in May, helping annualized revenue grow from a Sacra-estimated $10M at year-end 2025 to over $70M in July. With Grok tightening moderation via API, Venice’s “NSFW Higgsfield” now multiplexes Seedance (ByteDance), Kling, Wan (Alibaba), and other open source models jailbroken for minimum refusals, including Venice’s own fine-tuned Venice Uncensored, with video editing & references that preserve characters across scenes for more sophisticated creations.
- Sacra estimates Venice reached $150M in annualized revenue in September 2026, up 15x from $10M at year-end 2025, after raising a $65M Series A (Dragonfly) in July at a $1B valuation for a 14.3x multiple on its contemporaneous $70M revenue run rate. Compare to OpenArt at a Sacra-estimated $100M ARR in June 2026, up from $70M at year-end 2025 (+567% YoY in 2025), and Higgsfield at $1B in annualized revenue in September 2026, up 5x year-to-date, last valued at $5.4B in August for a 7.7x multiple on $700M in annualized revenue.
For more, check out this other research from our platform:
- Venice AI (dataset)
- Higgsfield (dataset)
- OpenArt (dataset)
- Runway (dataset)
- xAI (dataset)
- Kling (dataset)
- xAI vs OpenAI vs Anthropic
- Why SpaceX acquired xAI
- Truth Social: the GameStop for the alt-right
- Reddit: the $510M/year multimodal TikTok
- Higgsfield at $230M ARR
- $100M/year Pinterest for AI video
- Alex Mashrabov, CEO of Higgsfield, on orchestrating AI video models
- Coco Mao, CEO of OpenArt, on building the TikTok for AI video
- Cristóbal Valenzuela, CEO of Runway, on the state of generative AI in video
