AWS nabs white hot gen AI creation media startup fal, becomes its preferred cloud provider
Generative AI’s rapid transition from text-based chatbots to reliable media—expandable images, video, spatial 3D, and audio—has revealed an obvious bottleneck in the modern technology stack: infrastructure. Rendering pixels in real time requires an incredible amount of computation, and developers are increasingly struggling to manage different GPU clusters just to keep their applications online.
Enter fal, a media production platform that has become the connective tissue of 2.5 million developers worldwide, offering literally hundreds of leading AI image, video, and audio creation and editing models – from proprietary ones like OpenAI’s ChatGPT-Images-2.0 and Google Nano Banana Pro 2 to open its non-integrated API competitors.
Today, the San Francisco-based startup, which recently raised $4.5 billion following a $300 million round led by Sequoia Capital, announced that it has chosen Amazon Web Services (AWS) as its preferred cloud provider.
Although the financial terms of the deal were not disclosed publicly, the move reflects growth in the productive media space, shifting the focus from basic models to successful cuts for mass, commercial use.
“AWS existed for distribution and monetization, and for the use of AI in creative applications — to help designers, developers, and the creative community think about how to use AI responsibly, aggressively, and globally," said Samira Panah Bakhtiar, General Manager of Media, Entertainment, Games, and Sports at AWS, in an exclusive interview with VentureBeat.
A one-stop shop for Gen AI media that allows businesses to connect and choose the best model for their needs.
At its core, fal serves as a unified gateway to a rapidly growing productive AI ecosystem. Rather than forcing developers to provision their own servers, deal with latency issues, or bundle the burdens of an open source model, fal provides a single, unified API. With this API, users get instant access to over 1,000 production-ready AI models.
Think of it like Stripe or Plaid for productive media: removing complex back-end pipelines so developers can focus solely on the user experience.
It is a "plug-and-play" A solution that has already attracted independent creators and corporate giants alike, it powers productive workflows for businesses including Canva, Adobe, and Amazon MGM Studios.
“The workload of the production media requires a very different infrastructure layer, which can handle the same definition, rapid model iteration, and high production grade reliability,” said Gorkem Yurtseven, CTO and co-founder of fal, in a statement provided to VentureBeat.
Neither AWS nor fal specified what other cloud or GPU providers were using prior to their agreement together. Asked who he was using before AWS, Bakhtiar did not name the previous cloud or GPU provider, saying instead that the company now uses AWS services.
In a blog post, false Head of Computing Partnerships Emir Lise described AWS as providing a “global scale and reliability layer” for its existing serverless generative-media infrastructure – framing the partnership as about scalability, reliability and business scale rather than replacing the incumbent.
A public search revealed Tigris as a storage provider for fal – with Tigris saying that fal runs “a global array of GPUs across multiple clouds” – and an announcement from fal in Septemeber 2025 that it was available through the Google Cloud Marketplace, which allows customers to buy fake Google Cloud billing and governance, but that listing doesn’t mean Google Clouds infrastructure is Google infrastructure Cloud
99.99% uptime guaranteed?
In partnership with AWS, failure aims to combine its most advanced indexing engine with Amazon’s global reach to handle millions of daily API calls with 99.99% guaranteed uptime.
In addition, Bakhtiar said fake users can expect to see "faster visualization and performance, greater efficiency, more scalability, and seamless service continuity – all the things you can expect from partnering with the world’s largest, most widely adopted cloud."
Therefore, the main benefit of virtual users is better performance and reliability without changing the way they work: faster assumptions, more scalability, smoother progress, and access to production-ready AI models without managing their infrastructure.
At fal, partnerships make their platform powerful for creators, studios, and enterprise customers by supporting it with AWS’s security, global scale, and cloud infrastructure.
For AWS, it helps push cloud and AI deeper into creative production, not just distribution or monetization. It positions AWS as a key infrastructure partner for studios, media companies, developers, and individual creators building AI-powered content workflows.
Loads the GPU load
The partnership with AWS is designed to address the sheer physics and cost of providing productive media. By moving its operations to AWS, fal will be able to use a wide range of Amazon’s AI services, including the Bedrock platform, alongside custom-built silicon such as Trainium and Graviton processors.
"You don’t need to manage like GPU ships to use AI in creative tasks," Bakhtiar explained.
This is a key pain point for mass media production needs in 2026. Finding GPUs that perform well with the same understanding is both expensive and technically demanding.
By transferring that responsibility to AWS, fal ensures that developers can focus on their operations, without needing a dedicated DevOps team.
Bakhtiar also noted the powerful "network effect" of building on AWS. Because major studios and creative platforms (such as Adobe and Canva) are already entrenched in the AWS ecosystem, integrating Fal’s API into their existing pipelines becomes a constant effort.
Enterprise-grade security and compatibility with the creative speed of next-gen AI
For IT leaders and developers, fal architecture offers a distinct advantage in terms of licensing, security, and deployment.
Historically, using production models at the border meant accepting strict vendor lock-in from a single provider or trying to host open source models locally.
The latter requires significant overhead and forces businesses to navigate the minefield of different open source licenses (such as MIT, Apache 2.0, or restrictive non-commercial licenses).
fal overcomes this conflict by providing commercial API access to a select system of models. Developers simply pay for what they think they use.
In addition, the platform is compatible with SOC 2 and is clearly designed "business scale," meaning it meets the strict privacy and security standards required by highly regulated industries and large consumer platforms.
For major media organizations, this managed service approach allows them to test the latest technology securely, without the risk of exposing proprietary data or proprietary material.
It empowers devs and coders of vibe
The real impact of the fal platform, however, is best seen at the developer level. By democratizing high-quality infrastructure, fal created a new class of builders—often called "vibe codes"-creating complex, diverse applications outside of the traditional computer science background.
As Bakhtiar points out, access to these tools is critical "levels the playing field". Whether it’s an individual developer or hobbyist vibe coding a side project, or a fully funded editor or director pitching a blockbuster movie, the core technology is now the same, highly scalable, and ready for production.
“More creators – whether they’re full-fledged studios, indie brands, or individual content creators – will now have access to these tools, and will be able to punch above their weight as a result," Bakhtiar said, forming the partnership as a way to serve more users with false appreciation for the reliability of AWS servers and custom Trainium, Graviton and Inferentia chips.
The rollout of the enhanced AWS capabilities to virtual customers will occur in phases through 2026.
