Skip to content

Fal: $16 billion valuation for AI inference, or how much is cash worth?

3 min read
Fal: $16 млрд оценка за AI-инференс, или сколько стоит кеш? — illustration 1

San Francisco startup Fal, specializing in hosting and running AI models for video, images, and audio, has raised $1.3 billion in its latest funding round, reaching a valuation of $16 billion. This impressive growth highlights the rapid development of the AI inference market and its potential to generate significant profits. Investors, including Bessemer Venture Partners, Nvidia, Sound Ventures, and T. Rowe Price, demonstrate their confidence in the company’s future, whose valuation has doubled since April, when it was $8 billion, as reported by Bloomberg.

Revenue growth and profitability

Fal simultaneously announced Series E and F rounds, raising a total of $1.65 billion. The F round is driven by the growing demand for inference services — the process of running trained AI models to generate results. This is especially relevant for customers using open models.

Earlier this year, Fal exceeded $800 million in annual revenue, according to Bloomberg. Companies providing inference services, such as Baseten, Fireworks AI, and Modal, are also showing rapid growth and attracting significant investments with high valuations.

“Despite impressive revenue growth, gross margins in this segment remain low due to the high cost of purchasing or renting computing power,” experts familiar with the financial performance of such companies note.

Founders and market prospects

Fal was founded in 2021 by Burkay Gur, former head of machine learning at Coinbase and an Oracle engineer, and Görkem Yurtseven, formerly a developer at Amazon. Their project clearly demonstrates how CPM becomes profitable when scaling AI services.

For an arbitrageur, this is a signal: testing hypotheses cheaply might be precisely in the niche of AI content generation. the price per minute of AI video with an avatar or the cost of making one video unique can become new metrics for evaluating effectiveness.

Fal:  млрд оценка за AI-инференс, или сколько стоит кеш? — illustration 2

Frequently Asked Questions

What CPM is considered profitable in the context of AI inference?

A profitable CPM depends on the unit economics of your project. If the price per one ready-made clip generated by AI allows you to get an ROI, then the CPM will be profitable. The key here is price comparison: what’s cheaper — a banner or a blogger, or generating 100 AI videos?

Should one invest in AI services instead of traditional advertising?

Fact: Fal raised $1.3 billion. This speaks to the market’s belief in AI. For an arbitrageur, this means that a budget alternative to seeding or cheaper than influencers might be precisely in AI. The question is not “is AI worth it,” but “how much cash will it bring for every ruble spent.”

What are the risks associated with the low profitability of AI inference services?

Low profitability means dependence on volumes and the efficiency of computing resource utilization. For the user, this can be profitable, as CPM cheaper than 50 rubles becomes a reality. But for the service provider, this requires constant cost optimization and scaling.

Conclusion: how much money will it bring?

Fal’s growth to $16 billion confirms that the AI inference market is not just hype, but a source of real cash. For those looking for a cheap hypothesis test or a mass posting package price, it’s worth paying attention to the opportunities offered by AI services. We calculate ROI, compare price vs result, and draw conclusions. It’s not about emotions, it’s about numbers.

Don’t miss the opportunity to explore how AI inference can optimize your marketing campaigns and increase profits. Start analyzing your metrics today!


← Back to blog