Fractile Raises $720M to Challenge AI Chip Leaders with DRAM Breakthrough
Summarized by AI from reporting by TechCrunch AI, published under our editorial policy.
British AI chip startup Fractile secured $720 million in funding, positioning itself as a major player in the AI hardware race. The company's unique DRAM-based chip design aims to improve AI model performance and efficiency.

Key takeaways
- Fractile raised $720 million in a Series C funding round, valuing the company at over $7 billion.
- The company's unique DRAM-based chip design aims to improve AI model performance and efficiency.
- Fractile plans to ship its first chips and systems next year.
- The funding round was led by Thrive Capital, Founders Fund, Lightspeed Venture Partners, and Redpoint.
- Fractile's approach is designed to address the growing compute challenges faced by AI developers and researchers.
Fractile, a British AI chip startup, has raised $720 million in a Series C funding round led by Thrive Capital, Founders Fund, Lightspeed Venture Partners, and Redpoint. This significant investment, backed by Britain’s Sovereign AI Fund, makes Fractile one of the top-funded players in the race for purpose-built AI chips, alongside companies like Axelera, Etched, and OLIX. The funding underscores the growing importance of specialized hardware in the AI industry, as companies seek to reduce the cost and improve the performance of AI models in production.
Fractile's Unique Approach to AI Chips
Fractile's main differentiator is its hardware design, which revolves around DRAM (Dynamic Random-Access Memory). Unlike traditional AI chips, Fractile stacks DRAM on top of each chip, aiming to bypass the usual trade-off between memory and capacity. This innovative approach is designed to enhance the speed and efficiency of large language models (LLMs) and other AI applications. The company also offers a full-stack approach, including a machine learning compiler. This compiler is designed to maximize the speed at which LLMs are executed on Fractile's chips, setting the company apart from competitors who focus solely on hardware.
The company's approach is particularly relevant in the context of agentic AI, which involves performing longer chains of tasks with large models without compromising on speed. As AI models become more complex and demanding, the need for specialized hardware to reduce inference costs and improve performance has become critical. Fractile's DRAM-based chips aim to address these growing compute challenges faced by AI developers and researchers.
The Competitive Landscape of AI Chips
The AI chip market is highly competitive, with established players like NVIDIA, Google, and AMD, as well as numerous startups vying for a piece of the pie. NVIDIA's GPUs, for example, are widely used for AI training and inference, while Google's Tensor Processing Units (TPUs) are designed specifically for AI workloads. Fractile's entry into this market is significant because it offers a unique solution that could potentially outperform existing options in certain scenarios.
Fractile's full-stack approach, which includes both hardware and software components, is a key differentiator. By optimizing both the chip design and the machine learning compiler, Fractile aims to provide a more efficient and cost-effective solution for AI developers. This approach could make AI more accessible and affordable for a wider range of applications, from virtual assistants to complex AI-driven services.
Why Fractile's Funding Matters
The $720 million funding round values Fractile at over $7 billion, according to reports from the Financial Times and Reuters. This valuation underscores the growing importance of AI chips in the technology landscape. As AI models become more complex and demanding, the need for specialized hardware to reduce inference costs and improve performance has become critical.
Fractile's focus on DRAM-based chips is particularly relevant in the context of agentic AI, which involves performing longer chains of tasks with large models without compromising on speed. The company's approach aims to address the growing compute challenges faced by AI developers and researchers. By providing a more efficient and cost-effective solution, Fractile could help accelerate the adoption of AI across various industries.
What This Means for Everyday Users
While Fractile's chips are primarily targeted at developers and enterprises, the implications for everyday users are significant. More efficient AI chips can lead to faster, more responsive AI applications, from virtual assistants to complex AI-driven services. As AI becomes more integrated into daily life, the performance and efficiency of the underlying hardware will directly impact user experience.
For businesses, Fractile's chips could offer a cost-effective solution for deploying AI models in production. By reducing inference costs and improving performance, these chips could make AI more accessible and affordable for a wider range of applications. This could lead to more innovative AI-driven products and services, benefiting both businesses and consumers.
What You Can Do Today
While Fractile's chips are not yet available for purchase, the company plans to ship its first chips and systems next year. In the meantime, you can stay informed about the latest developments in AI hardware by following industry news and updates from Fractile. If you are a developer or enterprise looking to deploy AI models, keep an eye on Fractile's progress and consider how their chips could benefit your projects in the future.
Additionally, you can explore other AI hardware options currently available, such as NVIDIA's GPUs or Google's Tensor Processing Units (TPUs), to see how they compare to Fractile's upcoming offerings. Understanding the different options available can help you make informed decisions about the best hardware for your AI projects.
Frequently asked
- What makes Fractile's AI chips different from other AI hardware?
- Fractile's AI chips use a unique DRAM-based design, stacking DRAM on top of each chip to bypass the usual trade-off between memory and capacity. This approach aims to enhance the speed and efficiency of large language models (LLMs) and other AI applications.
- When will Fractile's chips be available for purchase?
- Fractile plans to ship its first chips and systems next year. The exact release date has not been specified.
- How does Fractile's funding round compare to other AI chip startups?
- Fractile's $720 million Series C funding round makes it one of the top-funded players in the race for purpose-built AI chips, alongside companies like Axelera, Etched, and OLIX.
- What is the valuation of Fractile after the funding round?
- According to reports from the Financial Times and Reuters, Fractile's valuation is over $7 billion.