San Francisco AI video platform Higgsfield raised $400 million led by DST Global, quadrupling its valuation to $5.4 billion just eight months after reaching unicorn status.
- DST Global led the August 2026 round, joined by Tribe Capital, Goldman Sachs Alternatives, Intel Capital, and six other investors.
- Annualized revenue reached $700 million; 30 million users across 238 countries generate 20 million content pieces monthly.
- The $5.4 billion price tag is a 4x jump from the $1.3 billion valuation set at Higgsfield's Series A in January 2026.
Lead
Higgsfield closed a $400 million Series B on August 17, 2026, setting a $5.4 billion valuation - a fourfold increase from the $1.3 billion price set eight months prior. DST Global led the round, with participation from Tribe Capital, Growth Equity at Goldman Sachs Alternatives, Smash Capital, Fifth Wall, Valor Capital, Intel Capital, Liberty Global Tech Ventures, Mirae Asset Capital, and NTT DOCOMO Ventures. The AI visual creation platform reported $700 million in annualized revenue and more than 30 million registered users at closing.
What Does Higgsfield Actually Build?
Higgsfield is not a single-model AI video generator. It aggregates more than 15 leading video and image models - including Sora 2, Google Veo 3.1, Kling 3.0, WAN 2.6, and Seedance 2.0 - under one subscription, positioning itself as a full-stack virtual production studio accessible from a browser or mobile app. Its Cinema Studio 3.5 applies an AI director layer that breaks a creative concept into individual shots. Marketing Studio converts a product URL into formatted ad content across multiple output specs. A Soul ID feature preserves character identity across separate generations.
Integrations with Photoshop and DaVinci Resolve let users generate footage directly inside professional editing tools without leaving familiar workflows.
What Drove a 4x Valuation Jump in Eight Months?
One product event explains most of the compression: Higgsfield's Supercomputer rollout in May 2026. In the three months that followed, users of its agentic products grew 42-fold, and monthly content generations reached 20 million. Annualized revenue climbed from roughly $200 million at the time of the Series A to $700 million - more than a 3x increase over the same period. The company now powers visual production for 390 of the Fortune 500, a customer base that carries both credibility and pricing leverage.
The speed of the valuation reset is notable. Most Series B rounds reflect 18-to-24 months of incremental progress. Here the gap was eight months. DST Global, which historically has entered consumer internet companies at late growth stages, does not lead rounds based on promises.
Founding and Prior Rounds
CEO Alex Mashrabov co-founded Higgsfield with CTO Yerzat Dulat and Mahi de Silva. Mashrabov previously ran Generative AI at Snap; before that, Snap acquired his startup AI Factory for $166 million in 2020. Higgsfield raised an $8 million seed from Menlo Ventures in April 2024, a $50 million Series A from GFT Ventures in September 2025 at a $1.0 billion valuation, and an $80 million Series A extension led by Accel in January 2026 at $1.3 billion before this round.
Competitive Context
The AI video generation market drew a dozen well-funded entrants between 2024 and 2026. OpenAI's Sora, Google's Veo series, and multiple venture-backed startups compete for creator and enterprise budgets. Higgsfield's structural bet is aggregation rather than model ownership. It does not need a single model to win the generation race; it needs the best available models under one roof with enough proprietary tooling - workflow automation, character consistency, ad formatting - to justify margin above raw API costs. The Fortune 500 customer count suggests that bet is converting at enterprise scale.
The risk is symmetrical: if foundation model providers restrict third-party access or build competing workflow layers, Higgsfield's aggregation advantage compresses quickly.
Outlook
Higgsfield plans to deploy the capital across global infrastructure, team expansion, and R&D. At $700 million in annualized revenue against a $5.4 billion valuation, the company trades at roughly 7.7x revenue - a multiple consistent with fast-growing SaaS businesses and more grounded than many AI-era benchmarks. The critical variable is whether the revenue trajectory set in the 90 days after the Supercomputer launch holds as that initial cohort effect normalizes. A post-launch acceleration that sustains through Q4 2026 would validate the valuation; one that fades would not.



