Funding · 13 Jul 2026 · 2 min read

PixVerse raises $439M for its video generation platform

The video generation startup's $2 billion valuation shows strong market demand for specialised AI products built on unique data strategies.

Pen-and-ink illustration: a single, large, polished stone is. For the story "PixVerse raises $439M for its video generation platform".
— Pen-and-ink illustration: a single, large, polished stone is. For the story "PixVerse raises $439M for its video generation platform". —

What happened

Singapore-based video startup PixVerse has closed a $439 million Series C extension, pushing its valuation past $2 billion. The company told TechCrunch the new funding will help expand its world model offering and grow its team. Investors in the round include Alibaba, Lollapalooza Capital, and Mirae Asset.

PixVerse currently has over 150 million registered users, with 15 million active monthly on its consumer product. The company was founded in 2023 by former ByteDance computer vision lead Wang Changhu and Jaden Xie.

How the room's reading it

The chatter around this raise centres on market validation for specialised AI products. PixVerse's co-founder Jaden Xie is framing their edge as superior data labelling — a technique pulled from his team's experience building TikTok's visual understanding tech. He's also bullish on the competitive landscape, telling TechCrunch that rivals like Meta and Tencent can't meet the quality bar.

However, industry watchers see a crowded field. Competitors include established players like Runway and Luma in the West, and new models from ByteDance and Kling AI in Asia. The consensus is that while the funding is a strong signal, PixVerse is entering a fiercely competitive race where differentiation will be key.

Sailfish's take

We're less interested in the valuation and more in the claim that data labelling is the core moat. If PixVerse is right, it’s a tough lesson for teams chasing scale with commodity data. It suggests the real work isn't just hoarding petabytes — it's the painstaking, domain-specific process of making that data useful. This is good news for focused, AI-native teams.

For builders, this reinforces a simple truth we've seen in our own work: a unique dataset, or a unique way of processing a public one, is still the most durable advantage you can build. The PixVerse raise isn't just a bet on video; it's a bet on applied data science over raw model scale.

Our take — your read?

Be the first to weigh in.

Sources
— END OF DISPATCH — Funding