Meta Just Signed a Massive 6GW AI Chip Deal With AMD After Nvidia Deal

Meta is dramatically expanding its artificial intelligence infrastructure. Just days after strengthening its partnership with Nvidia, the company has announced a major multi-year agreement with Advanced Micro Devices (AMD) to deploy up to 6 gigawatts (6GW) of AMD GPUs to power its next generation of AI data centres according to a report by CNBC.

This is not a minor procurement update. The deal is valued in the multi-billion-dollar range and involves the large-scale acquisition of advanced AI GPUs, placing it alongside major AI infrastructure investments by the likes of Microsoft and Tesla in recent years. That makes it one of the most notable AI hardware commitments, reflecting how aggressively companies are racing to secure the compute power needed as AI models become larger and more resource-intensive.

Meta has entered into a long-term strategic partnership with AMD to deploy up to 6GW of AMD Instinct GPUs. These graphics processing units (GPUs) will be used in Meta’s AI data centres to train and run advanced artificial intelligence models.

The announcement comes shortly after Meta expanded its agreement with Nvidia for millions of GPUs. In simple terms, Meta is not choosing one chip supplier over another. Instead, it is investing heavily in both.

This dual approach signals a deliberate effort to diversify its AI hardware supply chain.

The “6GW” figure refers to power capacity. A gigawatt equals one billion watts. So 6GW represents six billion watts of power capacity allocated to AI compute infrastructure.

To put that into perspective:

• It is comparable to the output of multiple large power stations.

• It represents an enormous data centre scale.

• It reflects the extreme energy demands of modern AI training models.

Artificial intelligence systems, particularly large language models and generative AI platforms, require immense computational resources. Training these systems involves processing vast quantities of data across thousands of GPUs simultaneously.

Meta’s 6GW commitment shows how serious it is about scaling its AI operations.

Meta has repositioned itself as an AI-first company. From content recommendations and advertising optimisation to generative AI tools and smart assistants, artificial intelligence now underpins much of its business.

Key areas driving Meta’s AI infrastructure growth include:

• Large language model development

• AI-powered advertising systems

• Content moderation tools

• AI assistants and chatbots

• Virtual and augmented reality development

AI hardware is the backbone of all these initiatives. Without high-performance GPUs, these systems cannot function efficiently.

By securing long-term GPU supply agreements with both AMD and Nvidia, Meta is safeguarding its ability to scale.

Nvidia currently dominates the AI chip market. Its GPUs are widely used for AI model training and high-performance computing.

However, relying on a single supplier carries risks:

• Supply chain bottlenecks

• Pricing pressures

• Limited negotiating leverage

• Slower innovation cycles

By partnering with AMD as well, Meta gains flexibility and competitive balance.

AMD’s Instinct GPU line has improved significantly in recent years. The company is positioning itself as a credible alternative in the AI accelerator market. Securing a multi-gigawatt deployment with Meta strengthens AMD’s standing in the AI chip race.

For Meta, diversification enhances resilience and bargaining power.

A 6GW GPU deployment raises another issue: energy consumption. AI data centres are power-intensive. Governments and regulators are increasingly scrutinising the environmental impact of large-scale computing facilities.

Meta and other technology firms face pressure to balance AI growth with sustainability goals. Renewable energy sourcing, data centre efficiency improvements, and advanced cooling technologies are becoming central to AI infrastructure planning.

This partnership therefore intersects not only with technology strategy, but also with environmental policy.

Meta’s broader strategy centres on building advanced AI systems that enhance its platforms and services. From social media algorithms to huge virtual environments, AI drives engagement and monetisation. 

The company has previously signalled plans to invest tens of billions of dollars in AI infrastructure. This AMD deal aligns with that commitment.

The scale of compute being secured suggests Meta expects AI demand to rise sharply over the coming years.

Meta’s 6GW AMD GPU deployment reflects three clear trends:

1. AI infrastructure spending is accelerating.

2. Chip supplier diversification is becoming standard practice.

3. Compute capacity is now a competitive advantage.

The artificial intelligence race is no longer just about algorithms. It is about hardware scale, energy access, and long-term supply agreements.

Companies that secure sufficient compute resources will be better positioned to train more powerful models and deploy advanced AI products.

RELATED: Meta is Introducing Premium Subscription Plans for Instagram, Facebook and WhatsApp — But Is It Worth Paying For?

Meta’s decision to deploy up to 6GW of AMD GPUs, shortly after expanding its Nvidia partnership, underscores the scale of today’s AI infrastructure race.