Large language models and the applications they power enable unprecedented opportunities for organizations to get deeper insights from their data reservoirs and to build entirely new classes of applications.
But with opportunities often come challenges.
Both on premises and in the cloud, applications that are expected to run in real time place significant demands on data center infrastructure to simultaneously deliver high throughput and low latency with one platform investment.
To drive continuous performance improvements and improve the return on infrastructure investments, NVIDIA regularly optimizes the state-of-the-art community models, including Meta's Llama, Google's Gemma, Microsoft's Phi and our own NVLM-D-72B, released just a few weeks ago.
Relentless Improvements Performance improvements let our customers and partners serve more complex models and reduce the needed infrastructure to host them. NVIDIA optimizes performance at every layer of the technology stack, including TensorRT-LLM, a purpose-built library to deliver state-of-the-art performance on the latest LLMs. With improvements to the open-source Llama 70B model, which delivers very high accuracy, we've already improved minimum latency performance by 3.5x in less than a year.
We're constantly improving our platform performance and regularly publish performance updates. Each week, improvements to NVIDIA software libraries are published, allowing customers to get more from the very same GPUs. For example, in just a few months' time, we've improved our low-latency Llama 70B performance by 3.5x.
NVIDIA has increased performance on the Llama 70B model by 3.5x. In the most recent round of MLPerf Inference 4.1, we made our first-ever submission with the Blackwell platform. It delivered 4x more performance than the previous generation.
This submission was also the first-ever MLPerf submission to use FP4 precision. Narrower precision formats, like FP4, reduces memory footprint and memory traffic, and also boost computational throughput. The process takes advantage of Blackwell's second-generation Transformer Engine, and with advanced quantization techniques that are part of TensorRT Model Optimizer, the Blackwell submission met the strict accuracy targets of the MLPerf benchmark.
Blackwell B200 delivers up to 4x more performance versus previous generation on MLPerf Inference v4.1's Llama 2 70B workload. Improvements in Blackwell haven't stopped the continued acceleration of Hopper. In the last year, Hopper performance has increased 3.4x in MLPerf on H100 thanks to regular software advancements. This means that NVIDIA's peak performance today, on Blackwell, is 10x faster than it was just one year ago on Hopper.
These results track progress on the MLPerf Inference Llama 2 70B Offline scenario over the past year. Our ongoing work is incorporated into TensorRT-LLM, a purpose-built library to accelerate LLMs that contain state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT-LLM is built on top of the TensorRT Deep Learning Inference library and leverages much of TensorRT's deep learning optimizations with additional LLM-specific improvements.
Improving Llama in Leaps and Bounds More recently, we've continued optimizing variants of Meta's Llama models, including versions 3.1 and 3.2 as well as model sizes 70B and the biggest model, 405B. These optimizations include custom quantization recipes, as well as efficient use of parallelization techniques to more efficiently split the model across multiple GPUs, leveraging NVIDIA NVLink and NVSwitch interconnect technologies. Cutting-edge LLMs like Llama 3.1 405B are very demanding and require the combined performance of multiple state-of-the-art GPUs for fast responses.
Parallelism techniques require a hardware platform with a robust GPU-to-GPU interconnect fabric to get maximum performance and avoid communication bottlenecks. Each NVIDIA H200 Tensor Core GPU features fourth-generation NVLink, which provides a whopping 900GB/s of GPU-to-GPU bandwidth. Every eight-GPU HGX H200 platform also ships with four NVLink Switches, enabling every H200 GPU to communicate with any other H200 GPU at 900GB/s, simultaneously.
Many LLM deployments use parallelism over choosing to keep the workload on a single GPU, which can have compute bottlenecks. LLMs seek to balance low latency and high throughput, with the optimal parallelization technique depending on application requirements.
For instance, if lowest latency is the priority, tensor parallelism is critical, as the combined compute performance of multiple GPUs can be used to serve tokens to users more quickly. However, for use cases where peak throughput across all users is prioritized, pipeline parallelism can efficiently boost overall server throughput.
The table below shows that tensor parallelism can deliver over 5x more throughput in minimum latency scenarios, whereas pipeline parallelism brings 50% more performance for maximum throughput use cases.
For production deployments that seek to maximize throughput within a given latency budget, a platform needs to provide the ability to effectively combine both techniques like in TensorRT-LLM.
Read the technical blog on boosting Llama 3.1 405B throughput to learn more about these techniques.
Different scenarios have different requirements, and parallelism techniques bring optimal performance for each of these scenarios. The Virtuous Cycle Over the lifecycle of our architectures, we deliver significant performance gains from ongoing software tuning and optimization. These improvements translate into additional value for customers who train and deploy on our platforms. They're able to create more capable models and applications and deploy their existing models using less infrastructure, enhancing th
Most recent headlines
05/01/2027
Worlds first 802.15.4ab-UWB chip verified by Calterah and Rohde & Schwarz to be ...
07/10/2026
Dalet, a leading technology and service provider for media-rich organizations, today announced the latest Long-Term Supported (LTS) release of Dalet Flex. Build...
06/09/2026
June 9 2026, 23:00 (PDT) Dolby and MagentaTV Bring Fans Closer to the FIFA Worl...
04/08/2026
Dalet, a leading technology and service provider for media-rich organizations, t...
16/07/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
16/07/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
16/07/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
16/07/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
16/07/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
16/07/2026
Zixi, a leader in live video delivery and workflow orchestration, and Comcast Technology Solutions (CTS) today announced a strategic new integration to deliver ...
16/07/2026
Blackmagic Design Cameras Capture Immersive Flight for New Cirrus App
Brie Clayton July 15, 2026
0 Comments
Leader in Personal Aviation taps URSA Cin...
16/07/2026
XenData Announces LTO Archive Appliances with both File and S3 Object Storage In...
16/07/2026
Together the semi-finals had a combined reach of 1.98 million unique individuals...
15/07/2026
The 2026 FIFA World Cup is a big deal for DAZN, which has distribution rights in three markets: Italy, Spain, and Japan. Each requires a different editorial app...
15/07/2026
As sports-media organizations manage rapidly growing content libraries and incre...
15/07/2026
HBS director of digital Johannes Franken talks SVG Europe through the operation built to serve FIFA World Cup 2026 rights holders remotely from Stratford, east ...
15/07/2026
IBC has unveiled the conference program for IBC2026, with this year's event ...
15/07/2026
Software-defined wideband RF platform supports 96-kHz monitoring, streamlined fr...
15/07/2026
New AI-powered module enables broadcasters and content owners to extract dialogu...
15/07/2026
Weeklong gathering at EBU headquarters will test new IPMX capabilities and expand certified product ecosystem...
15/07/2026
Force Tech has announced the appointment of Bryan Adams, CTS, as Vice President of Integration. Adams will oversee AV and broadcast integration project delivery...
15/07/2026
USA Sports has announced an exclusive English-language agreement with the Bundes...
15/07/2026
ITV Sport's FIFA World Cup 2026 coverage has been produced from one of the m...
15/07/2026
Program will provide professional camera kits, training, and technical support to participating local newsrooms...
15/07/2026
San Francisco church becomes first U.S. deployment of the 2 M/E production switcher...
15/07/2026
Former ESPN and NFL Network executive to advise startup league on media, distribution, technology, and fan engagement...
15/07/2026
The Houston-based RSN is the live-game home of the Astros and Rockets...
15/07/2026
Launched alongside new FreqSoft installer
Designed to bridge the gap between hardware and software, Freqport's FreqInOut FO1 unit provides a simple way ...
15/07/2026
Save up to 55% on software until 23 July 2026
Expressive E have just launched their Summer Offer sale, with discounts of up to 55% now applied across their ...
15/07/2026
Real-time stem-separation tool updated
zplane have recently released a new and improved version of their real-time stem-separation plug-in which they say ma...
15/07/2026
A new benchmark in bass enhancement
Described as a new benchmark in bass enhancement , Baby Audio's latest release has been designed to augment the lo...
15/07/2026
DMC Production's AR 1 and AR 2 units reverse the conventional OB workflow, keeping all production-critical processing at the venue while operators work remo...
15/07/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
15/07/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
15/07/2026
GigaCloud, a leading Ukrainian cloud service provider, and Cubbit, the first geo-distributed cloud storage enabler, today announced a new Business Alliance Part...
15/07/2026
SMPTE, the global society for media professionals, technologists, and engineers, has announced the launch of the SMPTE 110th Anniversary Campaign, a fundraising...
15/07/2026
Studio Technologies, a leading manufacturer of high-quality audio, video, and fiber-optic solutions, unveils the new StudioComm Model 762 Stereo Level Control, ...
15/07/2026
COW Jobs: Producer, Virtual Events - Hybrid, Full Time, Salaried
Brie Clayton July 15, 2026
0 Comments
Producer, Virtual Events (Hybrid | Full-time | ...
15/07/2026
Brazilian media organization builds nationwide CDN with Broadpeak's high-performance software for low-latency streaming and new content distribution service...
15/07/2026
DHD announces a new utility designed to increase the efficiency of its audio mixing systems during initial installation and subsequent configuration.
Anyone w...
15/07/2026
BCNEXXT Brings Next-Generation Vipe to IBC2026, Breaking the 1:1 Channel-to-Server Ratio New innovations liberate broadcasters from legacy playout infrastructur...
15/07/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
15/07/2026
Share
Copy link
Facebook
X
Linkedin
Bluesky
Email...
15/07/2026
General-purpose robots and autonomous machines are moving from research labs to ...
15/07/2026
Wednesday 15 July 2026
Screeners and press pack now available: Gomorrah: The Or...
15/07/2026
Trinity's Treasures features Ruth Negga, Eleanor McEvoy, and David Norris, a...
15/07/2026
Join geographer Dr Susan Hegarty, Engineer Tim Joyce and Architect Orla Murphy a...
15/07/2026
Home to leading manufacturers, robotics pioneers, infrastructure builders and iconic gaming companies, of course, Japan is one of the world's centers of AI ...
14/07/2026
South end zone videoboard, cloud-based control system will be ready for 2026 football season...
Large language models and the applications they power enable unprecedented opportunities for organizations to get deeper insights from their data reservoirs and to build entirely new classes of applications.
