Sony Pixel Power calrec Sony

How Scaling Laws Drive Smarter, More Powerful AI

12/02/2025

Just as there are widely understood empirical laws of nature - for example, what goes up must come down, or every action has an equal and opposite reaction - the field of AI was long defined by a single idea: that more compute, more training data and more parameters makes a better AI model.

However, AI has since grown to need three distinct laws that describe how applying compute resources in different ways impacts model performance. Together, these AI scaling laws - pretraining scaling, post-training scaling and test-time scaling, also called long thinking - reflect how the field has evolved with techniques to use additional compute in a wide variety of increasingly complex AI use cases.

The recent rise of test-time scaling - applying more compute at inference time to improve accuracy - has enabled AI reasoning models, a new class of large language models (LLMs) that perform multiple inference passes to work through complex problems, while describing the steps required to solve a task. Test-time scaling requires intensive amounts of computational resources to support AI reasoning, which will drive further demand for accelerated computing.

What Is Pretraining Scaling? Pretraining scaling is the original law of AI development. It demonstrated that by increasing training dataset size, model parameter count and computational resources, developers could expect predictable improvements in model intelligence and accuracy.

Each of these three elements - data, model size, compute - is interrelated. Per the pretraining scaling law, outlined in this research paper, when larger models are fed with more data, the overall performance of the models improves. To make this feasible, developers must scale up their compute - creating the need for powerful accelerated computing resources to run those larger training workloads.

This principle of pretraining scaling led to large models that achieved groundbreaking capabilities. It also spurred major innovations in model architecture, including the rise of billion- and trillion-parameter transformer models, mixture of experts models and new distributed training techniques - all demanding significant compute.

And the relevance of the pretraining scaling law continues - as humans continue to produce growing amounts of multimodal data, this trove of text, images, audio, video and sensor information will be used to train powerful future AI models.

Pretraining scaling is the foundational principle of AI development, linking the size of models, datasets and compute to AI gains. Mixture of experts, depicted above, is a popular model architecture for AI training. What Is Post-Training Scaling? Pretraining a large foundation model isn't for everyone - it takes significant investment, skilled experts and datasets. But once an organization pretrains and releases a model, they lower the barrier to AI adoption by enabling others to use their pretrained model as a foundation to adapt for their own applications.

This post-training process drives additional cumulative demand for accelerated computing across enterprises and the broader developer community. Popular open-source models can have hundreds or thousands of derivative models, trained across numerous domains.

Developing this ecosystem of derivative models for a variety of use cases could take around 30x more compute than pretraining the original foundation model.

Developing this ecosystem of derivative models for a variety of use cases could take around 30x more compute than pretraining the original foundation model.

Post-training techniques can further improve a model's specificity and relevance for an organization's desired use case. While pretraining is like sending an AI model to school to learn foundational skills, post-training enhances the model with skills applicable to its intended job. An LLM, for example, could be post-trained to tackle a task like sentiment analysis or translation - or understand the jargon of a specific domain, like healthcare or law.

The post-training scaling law posits that a pretrained model's performance can further improve - in computational efficiency, accuracy or domain specificity - using techniques including fine-tuning, pruning, quantization, distillation, reinforcement learning and synthetic data augmentation.

Fine-tuning uses additional training data to tailor an AI model for specific domains and applications. This can be done using an organization's internal datasets, or with pairs of sample model input and outputs.

Distillation requires a pair of AI models: a large, complex teacher model and a lightweight student model. In the most common distillation technique, called offline distillation, the student model learns to mimic the outputs of a pretrained teacher model.

Reinforcement learning, or RL, is a machine learning technique that uses a reward model to train an agent to make decisions that align with a specific use case. The agent aims to make decisions that maximize cumulative rewards over time as it interacts with an environment - for example, a chatbot LLM that is positively reinforced by thumbs up reactions from users. This technique is known as reinforcement learning from human feedback (RLHF). Another, newer technique, reinforcement learning from AI feedback (RLAIF), instead uses feedback from AI models to guide the learning process, streamlining post-training efforts.

Best-of-n sampling generates multiple outputs from a language model and selects the one with the highest reward score based on a reward model. It's often used to improve an AI's outputs without modifying model parameters, offering an alternative to fine-tuning with reinforcement learning.

Search methods explore a range of potential decision paths before selecting a final output. This post-training technique can iteratively improve the model's responses
LINK: https://blogs.nvidia.com/blog/ai-scaling-laws/...
See more stories from nvidia

More from Nvidia

28/04/2025

Oracle Cloud Infrastructure Deploys Thousands of NVIDIA Blackwell GPUs for Agentic AI and Reasoning Models

Oracle has stood up and optimized its first wave of liquid-cooled NVIDIA GB200 N...

24/04/2025

NVIDIA Research at ICLR - Pioneering the Next Wave of Multimodal Generative AI

Advancing AI requires a full-stack approach, with a powerful foundation of computing infrastructure - including accelerated processors and networking technologi...

24/04/2025

All Roads Lead Back to Oblivion: Bethesda's The Elder Scrolls IV: Oblivion Remastered' Arrives on GeForce NOW

Get the controllers ready and clear the calendar - it's a jam-packed GFN Thu...

23/04/2025

How the Economics of Inference Can Maximize AI Value

As AI models evolve and adoption grows, enterprises must perform a delicate balancing act to achieve maximum value. That's because inference - the process ...

23/04/2025

Capital One Banks on AI for Financial Services

Financial services has long been at the forefront of adopting technological innovations. Today, generative AI and agentic systems are redefining the industry, f...

23/04/2025

Project G-Assist Plug-In Builder Lets Anyone Customize AI on GeForce RTX AI PCs

AI is rapidly reshaping what's possible on a PC - whether for real-time image generation or voice-controlled workflows. As AI capabilities grow, so does the...

23/04/2025

Enterprises Onboard AI Teammates Faster With NVIDIA NeMo Tools to Scale Employee Productivity

An AI agent is only as accurate, relevant and timely as the data that powers it....

22/04/2025

Keeping AI on the Planet: NVIDIA Technologies Make Every Day About Earth Day

Whether at sea, land or in the sky - even outer space - NVIDIA technology is helping research scientists and developers alike explore and understand oceans, wil...

22/04/2025

Chill Factor: NVIDIA Blackwell Platform Boosts Water Efficiency by Over 300x

Traditionally, data centers have relied on air cooling - where mechanical chillers circulate chilled air to absorb heat from servers, helping them maintain opti...

22/04/2025

Making Brain Waves: AI Startup Speeds Disease Research With Lab in the Loop

About 15% of the world's population - over a billion people - are affected by neurological disorders, from commonly known diseases like Alzheimer's and ...

17/04/2025

AI Bites Back: Researchers Develop Model to Detect Malaria Amid Venezuelan Gold Rush

Gold prospecting in Venezuela has led to a malaria resurgence, but researchers h...

17/04/2025

Spring Into Action With 11 New Games on GeForce NOW

As the days grow longer and the flowers bloom, GFN Thursday brings a fresh lineup of games to brighten the week. Dive into thrilling hunts and dark fantasy adv...

16/04/2025

Isomorphic Labs Rethinks Drug Discovery With AI

Isomorphic Labs is reimagining the drug discovery process with an AI-first approach. At the heart of this work is a new way of thinking about biology. Max Jade...

16/04/2025

Into the Omniverse: How Digital Twins Are Scaling Industrial AI

Editor's note: This post is part of Into the Omniverse, a series focused on how developers, 3D practitioners, and enterprises can transform their workflows ...

15/04/2025

Math Test? No Problems: NVIDIA Team Scores Kaggle Win With Reasoning Model

The final days of the AI Mathematical Olympiad's latest competition were a transcontinental relay for team NVIDIA. Every evening, two team members on oppos...

15/04/2025

Everywhere, All at Once: NVIDIA Drives the Next Phase of AI Growth

Every company and country wants to grow and create economic opportunity - but they need virtually limitless intelligence to do so. Working with its ecosystem pa...

15/04/2025

Thousands of NVIDIA Grace Blackwell GPUs Now Live at CoreWeave, Propelling Development for AI Pioneers

CoreWeave today became one of the first cloud providers to bring NVIDIA GB200 NV...

14/04/2025

NVIDIA to Manufacture American-Made AI Supercomputers in US for First Time

NVIDIA is working with its manufacturing partners to design and build factories that, for the first time, will produce NVIDIA AI supercomputers entirely in the ...

11/04/2025

Beyond CAD: How nTop Uses AI and Accelerated Computing to Enhance Product Design

As a teenager, Bradley Rothenberg was obsessed with CAD: computer-aided design software. Before he turned 30, Rothenberg channeled that interest into building ...

10/04/2025

NVIDIA Celebrates Partners of the Year Advancing AI in Europe, Middle East and Africa

NVIDIA this week recognized the contributions of partners in Europe, the Middle ...

10/04/2025

(AI)ways a Cut Above: GeForce RTX 50 Series Accelerates New DaVinci Resolve 20 Studio Video Editing Software

As AI-powered tools continue to evolve, NVIDIA GeForce RTX 50 Series and NVIDIA ...

10/04/2025

Myth and Mystery Await: GeForce NOW Brings South of Midnight' to the Cloud at Launch

Get ready to explore the Deep South. South of Midnight, the action-adventure gam...

09/04/2025

Black Women in Artificial Intelligence' Founder Talks AI Education and Empowerment

Necessity is the mother of invention. And sometimes, what a person really needs ...

09/04/2025

NVIDIA Brings Agentic AI Reasoning to Enterprises With Google Cloud

NVIDIA is collaborating with Google Cloud to bring agentic AI to enterprises seeking to locally harness the Google Gemini family of AI models using the NVIDIA B...

06/04/2025

National Robotics Week - Latest Physical AI Research, Breakthroughs and Resources

This National Robotics Week, NVIDIA highlighted the pioneering technologies that...

03/04/2025

Nintendo Switch 2 Leveled Up With NVIDIA AI-Powered DLSS and 4K Gaming

The Nintendo Switch 2, unveiled April 2, takes performance to the next level, powered by a custom NVIDIA processor featuring an NVIDIA GPU with dedicated RT Cor...

03/04/2025

NVIDIA Showcases Real-Time AI and Intelligent Media Workflows at NAB

Real-time AI is unlocking new possibilities in media and entertainment, improving viewer engagement and advancing intelligent content creation. At NAB Show, a...

03/04/2025

From Browsing to Buying: How AI Agents Enhance Online Shopping

Editor's note: This post is part of the AI On blog series, which explores the latest techniques and real-world applications of agentic AI, chatbots and copi...

03/04/2025

No Foolin': GeForce NOW Gets 21 Games in April

GeForce NOW isn't fooling around. This month, 21 games are joining the cloud gaming library of over 2,000 titles. Whether chasing epic adventures, testing ...

02/04/2025

Speed Demon: NVIDIA Blackwell Takes Pole Position in Latest MLPerf Inference Results

In the latest MLPerf Inference V5.0 benchmarks, which reflect some of the most c...

02/04/2025

NVIDIA's Jacob Liberman on Bringing Agentic AI to Enterprises

AI is rapidly transforming how organizations solve complex challenges. The early stages of enterprise AI adoption focused on using large language models to cre...

02/04/2025

NVIDIA GeForce RTX 50 Series Accelerates Adobe Premiere Pro and Media Encoder's 4:2:2 Color Sampling

Video editing workflows are getting a lot more colorful. Adobe recently announc...

31/03/2025

Industrial Ecosystem Adopts Mega NVIDIA Omniverse Blueprint to Train Physical AI in Digital Twins

Advances in physical AI are enabling organizations to embrace embodied AI across...

27/03/2025

The Dream Life Awaits: Play inZOI' on GeForce NOW Anytime, Anywhere

A new resident is moving into the cloud - KRAFTON's inZOI joins the 2,000+ games in the GeForce NOW cloud gaming library. Plus, members can get ready for a...

26/03/2025

Buzz Solutions Uses Vision AI to Supercharge the Electric Grid

The reliability of the electric grid is critical. From handling demand surges and evolving power needs to preventing infrastructure failures that can cause wil...

25/03/2025

NVIDIA NIM Microservices Now Available to Streamline Agentic Workflows on RTX AI PCs and Workstations

Generative AI is unlocking new capabilities for PCs and workstations, including ...

20/03/2025

EPRI, NVIDIA and Collaborators Launch Open Power AI Consortium to Transform the Future of Energy

The power and utilities sector keeps the lights on for the world's populatio...

20/03/2025

Assassin's Creed Shadows' Emerges From the Mist on GeForce NOW

Time to sharpen the blade. GeForce NOW brings a legendary addition to the cloud: Ubisoft's highly anticipated Assassin's Creed Shadows is now available ...

19/03/2025

Innovation to Impact: How NVIDIA Research Fuels Transformative Work in AI, Graphics and Beyond

The roots of many of NVIDIA's landmark innovations - the foundational techno...

19/03/2025

NVIDIA Blackwell Powers Real-Time AI for Entertainment Workflows

AI has been shaping the media and entertainment industry for decades, from early recommendation engines to AI-driven editing and visual effects automation. Real...

19/03/2025

NVIDIA Honors Americas Partners Advancing Agentic and Physical AI

NVIDIA this week recognized 14 partners leading the way across the Americas for their work advancing agentic and physical AI across industries. The 2025 Americ...

18/03/2025

AI on the Menu: Yum! Brands and NVIDIA Partner to Accelerate Restaurant Industry Innovation

The quick-service restaurant industry is a marvel of modern logistics, where spe...

18/03/2025

Telecom Leaders Call Up Agentic AI to Improve Network Operations

Global telecommunications networks can support millions of user connections per day, generating more than 3,800 terabytes of data per minute on average. That m...

18/03/2025

NVIDIA Aerial Expands With New Tools for Building AI-Native Wireless Networks

The telecom industry is increasingly embracing AI to deliver seamless connections - even in conditions of poor signal strength - while maximizing sustainability...

18/03/2025

From AT&T to the United Nations, AI Agents Redefine Work With NVIDIA AI Enterprise

AI agents are transforming work, delivering time and cost savings by helping peo...

18/03/2025

Full Steam Ahead: NVIDIA-Certified Program Expands to Enterprise Storage for Faster AI Factory Deployment

AI deployments thrive on speed, data and scale. That's why NVIDIA is expandi...

18/03/2025

NVIDIA Accelerated Quantum Research Center to Bring Quantum Computing Closer

As quantum computers continue to develop, they will integrate with AI supercomputers to form accelerated quantum supercomputers capable of solving some of the w...

18/03/2025

AI Factories Are Redefining Data Centers and Enabling the Next Era of AI

AI is fueling a new industrial revolution - one driven by AI factories. Unlike traditional data centers, AI factories do more than store and process data - the...

18/03/2025

Accelerating AI Development With NVIDIA RTX PRO Blackwell Series GPUs and NVIDIA NIM Microservices for RTX

As generative AI capabilities expand, NVIDIA is equipping developers with the to...

17/03/2025

Explaining Tokens - the Language and Currency of AI

Under the hood of every AI application are algorithms that churn through data in their own language, one based on a vocabulary of tokens. Tokens are tiny units...