
If you're working at the cutting edge of AI and high-performance computing (HPC), NVIDIA's H200 and B200 GPUs are likely already very much on your radar. They're each designed to handle the massive compute demands of datacenter-class workloads – such as training trillion-parameter models and running high-speed inference at scale.

While both GPUs deliver incredible power, it's crucial to understand the differences between the two in order to make the right choice for your needs.

Today, we'll explore how the **H200** and **B200** measure up against each other and how each fits into real-world deployments. Which one is right for you? Let's take a look.

### NVIDIA H200: Hopper-Based Heavyweight

Building on the foundation laid by the H100, the NVIDIA H200 is based on the same Hopper architecture while bringing substantial improvements in memory capacity and bandwidth. This means fewer bottlenecks, an advantage that becomes increasingly important as you move further into enterprise scale.

The H200 nearly doubles the memory of its predecessor, boasting **141 GB of HBM3e memory** and **4.8 TB/s of bandwidth** – delivering about **1.4X faster** data access. MLPerf benchmarks using Llama 2 70B demonstrated the H200's power: it reached over 31,000 tokens per second, about **45% faster** than the H100.

With the same **fourth-gen NVLink**, **PCIe Gen5**, **MIG support**, and full Tensor Core and Transformer Engine capabilities, the H200 is a solid next step for existing Hopper-based deployments. You don't have to rethink your entire infrastructure; you just get more horsepower.

To sum up, here are some of the most notable features of the H200:

-   **141 GB of HBM3e memory at 4.8 TB/s** – almost double the memory of the H100 with 1.4X the bandwidth.

-   **Up to 1.9X improved LLM inference performance**, with more tokens per second and higher throughput for gen AI workloads compared to its predecessor.

-   **Lower energy use and total cost of ownership** (TCO) over the H100 thanks to improved energy efficiency and thermal management.

-   **An upgrade path exists on qualified HGX/MGX platforms**, so that little to no reconfiguration is required.

In short, the H200 represents the very best of the Hopper-based lineup. When it was introduced in late 2024, it set a new standard for memory-intensive and high-throughput workloads in datacenter GPUs – and it's still one of the most powerful GPUs on the market today.

The B200, on the other hand, is an entire generational leap forward.

### NVIDIA B200: The New Generation

Built on NVIDIA's advanced Blackwell architecture, the B200 features **192 GB of ultra-fast HBM3e memory** with **8.0 TB/s of memory bandwidth.** Those are some staggering numbers, even compared to the H200's respectable 141 GB at 4.8 TB/s.

The Blackwell architecture also introduces **fifth-gen Tensor Cores** and **dual Transformer Engines**, which are particularly useful for AI workloads with long context windows and heavy token parallelism. The B200 scales much more efficiently than the H200 for the latest and most advanced model classes.

In fact, scaling is a major area where the B200 stands out. With **fifth-gen** **NVLink**, its inter-GPU bandwidth reaches **1.8 TB/s** within a node – compared to Hopper's 900 GB/s – with cross-node scaling supported via the NVLink Switch System.

Of course, this may not matter to you very much if you're only using a single card. At cluster scale, however, or in multi-GPU systems like the DGX B200 (which connects eight B200 GPUs together), the improvements in distributed training efficiency really make a difference.

The B200 truly sits at the frontier of performance. Its raw capacity and speed can support next-gen inference pipelines and AI workloads at hyperscale. According to NVIDIA, it delivers **3X the training performance** and **15X the inference performance** of the previous generation.

These gains do come at a cost in power draw. The B200 has a max thermal design power (TDP) of **1000 W** compared to the H200 at 700 W – which means it also requires far more robust cooling solutions. But the extra power consumption may just be worth it for the boost in performance.

To recap, a few advantages of the B200 include:

-   **192 GB of HBM3e memory at 8.0 TB/s** – ultra-high capacity that's well suited for trillion-parameter models and allows training of AI models _entirely in GPU memory_.

-   **Fifth-gen Tensor Cores and second-gen Transformer Engines** for higher efficiency in FP8, FP16, and mixed precision performance, especially in long-context and multi-modal workloads.

-   **NVLink 5 interconnect** with up to **1.8 TB/s** GPU-to-GPU bandwidth for multi-node scaling.

-   **Fine-grain micro-tensor scaling** that optimizes performance and accuracy, enabling 4-bit floating point (**FP4**) AI.

The B200 was built to shatter performance records. As you can imagine, it has an eyewatering price tag to match. Nonetheless, it can be helpful to compare the B200 and H200 based on specs alone.

### H200 vs. B200: Key Specs Compared

Here's a side-by-side look at some key features of the H200 and B200:

<table class="w-full border-collapse border border-gray-300">
    <thead>
        <tr class="bg-gray-100">
            <th class="border border-gray-300 px-4 py-2 text-left font-semibold">
                Feature
            </th>
            <th class="border border-gray-300 px-4 py-2 text-left font-semibold">
                H200
            </th>
            <th class="border border-gray-300 px-4 py-2 text-left font-semibold">
                B200
            </th>
        </tr>
    </thead>
    <tbody>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Architecture
            </td>
            <td class="border border-gray-300 px-4 py-2">Hopper</td>
            <td class="border border-gray-300 px-4 py-2">Blackwell</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Memory
            </td>
            <td class="border border-gray-300 px-4 py-2">141 GB HBM3e</td>
            <td class="border border-gray-300 px-4 py-2">192 GB HBM3e</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Memory Bandwidth
            </td>
            <td class="border border-gray-300 px-4 py-2">4.8 TB/s</td>
            <td class="border border-gray-300 px-4 py-2">8 TB/s</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                CUDA Cores
            </td>
            <td class="border border-gray-300 px-4 py-2">16,896</td>
            <td class="border border-gray-300 px-4 py-2">16,896 x2</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Tensor Cores
            </td>
            <td class="border border-gray-300 px-4 py-2">528 (4th Gen)</td>
            <td class="border border-gray-300 px-4 py-2">528 x2 (5th Gen)</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Boost Clock
            </td>
            <td class="border border-gray-300 px-4 py-2">1.98 GHz</td>
            <td class="border border-gray-300 px-4 py-2">1.98 GHz</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Transformer Engine
            </td>
            <td class="border border-gray-300 px-4 py-2">Single</td>
            <td class="border border-gray-300 px-4 py-2">Single (2nd gen)</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Form Factor
            </td>
            <td class="border border-gray-300 px-4 py-2">SXM</td>
            <td class="border border-gray-300 px-4 py-2">SXM</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Interconnect
            </td>
            <td class="border border-gray-300 px-4 py-2">
                NVLink 4 (900 GB/s) per GPU
            </td>
            <td class="border border-gray-300 px-4 py-2">
                NVLink 5 (1.8 TB/s) per GPU
            </td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Precision Formats
            </td>
            <td class="border border-gray-300 px-4 py-2">
                FP8, FP16/BF16, TF32, FP32, FP64, INT8
            </td>
            <td class="border border-gray-300 px-4 py-2">
                FP4, FP6, FP8, FP16/BF16, TF32, FP32, FP64, INT8
            </td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Multi-Instance GPUs
            </td>
            <td class="border border-gray-300 px-4 py-2">Up to 7</td>
            <td class="border border-gray-300 px-4 py-2">Up to 7</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Max Thermal Design Power (TDP)
            </td>
            <td class="border border-gray-300 px-4 py-2">700 W</td>
            <td class="border border-gray-300 px-4 py-2">1000 W</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Recommended Power Supply
            </td>
            <td class="border border-gray-300 px-4 py-2">1100 W</td>
            <td class="border border-gray-300 px-4 py-2">1400 W</td>
        </tr>
    </tbody>
</table>

Choosing the right GPU often comes down to more than just specs, however.

### Use Cases: Which GPU Should You Choose?

While both the H200 and B200 deliver game-changing performance capabilities, they each have distinct advantages depending on your needs.

#### You should consider the H200 if you prefer:

-   **A high-performance GPU** ideal for fine-tuning LLMs and high-throughput inference at the enterprise level without getting into exascale territory.

-   **A proven workhorse** that balances strong performance with lower TDP and cost.

-   **Being able to seamlessly upgrade** from existing Hopper deployments without overhauling infrastructure.

#### Or the B200 might be the right choice if you need:

-   **To push model boundaries** and handle long-context and/or multi-modal foundation model training at scale.

-   **The absolute maximum throughput and efficiency** for multi-GPU or multi-node clusters.

-   **A future-proof investment** for infrastructure built to support the next generation of cutting-edge AI and AGI.

### Conclusion

At **Vast.ai**, we know that the choice between GPUs like the H200 and B200 can depend on a lot more than just performance. Access and cost are important factors, as well.

The H200 and B200 represent the very best of NVIDIA's current lineup, and they're among the most powerful and expensive GPUs in the world today. Buying them outright is undoubtedly out of reach for most teams.

That's what we're here for. With Vast, you can rent high-end GPUs like the [A100](https://vast.ai/article/H100-vs-A100-Comparing-two-Powerhouse-GPUs), [H100](https://vast.ai/article/nvidia-h100-vs-l40s-power-meets-versatility), [H200](https://vast.ai/article/nvidia-h100-vs-h200-two-hopper-based-heavyweights), and soon the B200, with on-demand pricing and spot instances – and pay only for what you use.

On average, **Vast users save 5–6X on GPU compute** compared to traditional cloud providers. You'll find a variety of options on our platform to fit your needs, so you can focus on what really matters: your work, not your budget.

Explore Vast today and enjoy enterprise-grade compute without enterprise-level costs.


The H200 and B200 GPUs represent the top end of NVIDIA's current lineup for AI and high-performance computing (HPC). Both are designed for serious, large-scale workloads – from training foundation models to running complex scientific simulations.

NVIDIA H200 vs. B200: Comparing Datacenter-Grade Accelerators


The GeForce RTX 4090 and the A100 both sit at the high end of NVIDIA's GPU lineup, but they're built for very different worlds.

The **RTX 4090** is a powerful consumer-grade GPU, designed for ultra high-performance gaming, professional creative work, and even some entry-level AI workloads. It's surprisingly capable across a variety of tasks, especially for its price point.

The **A100**, on the other hand, isn't aimed at consumers at all. It's an enterprise-grade GPU built for data centers, research labs, and AI teams running large-scale model training, inference, and simulation workloads – designed to move serious data at serious speed.

In this post, we'll look at how these two machines compare with each other, and why you might choose one over the other based on your specific needs.

### NVIDIA GeForce RTX 4090: A Workhorse with Range

When it launched in September 2022 as the top-tier consumer GPU of its generation, the RTX 4090 quickly became the go-to choice for 4K gaming, 3D rendering, and AI-enhanced creative workflows.

Built on NVIDIA's **Ada Lovelace** architecture and powered by the **AD102 chip**, the RTX 4090 introduced significant upgrades in ray tracing and Tensor Core acceleration, and delivered notable performance gains with **DLSS 3** in a variety of games.

While it's no longer the flagship GPU of NVIDIA's GeForce line (that title now goes to the RTX 5090, which you can read about [here](https://vast.ai/article/nvidia-geforce-rtx-5090-specs-everything-you-need-to-know)), the RTX 4090 definitely still holds its own. It remains one of the most capable and well-rounded GPUs available at its price point for consumers and technical teams alike.

#### Some of the RTX 4090's key specs and features include:

-   **24 GB of GDDR6X memory** with **1,008 GB/s bandwidth** on a **384-bit** interface.  

-   **16,384 CUDA cores** for high-parallel throughput.  

-   **128 third-gen RT cores** and **512 fourth-gen Tensor cores** for advanced ray tracing and AI workloads.  

-   **DLSS 3** support for real-time upscaling and performance boosts in many games.  

-   **450 watts** of thermal design power (TDP), requiring an **850-watt power supply** due to its high processing capabilities.

If you're looking for serious performance in a desktop form factor, the RTX 4090 continues to deliver. It may not be the newest card on the market, but it remains a workhorse that gets the job done across a wide range of applications.

As capable as the RTX 4090 is, it's not built for everything. If your workloads go beyond what a high-end consumer GPU can handle, the A100 just might bring the compute power you need.

### NVIDIA A100: A Powerhouse with Purpose

The A100 is built for scale. Based on NVIDIA's Ampere architecture and powered by the **GA100 chip**, it's designed to tackle massive workloads – like foundational AI model training – thanks to its raw throughput, high-bandwidth memory, and ability to scale across multi-GPU environments.

When the A100 launched in June 2021, it introduced **third-gen Tensor Cores**, support for **TF32** precision, and **Multi-Instance GPU** (MIG) capabilities – enabling flexible resource partitioning and efficient parallel processing across shared environments.

While some of its core specs might look modest compared to the RTX 4090, the A100 delivers superior performance in AI and data-intensive workloads because it's optimized where it counts: memory bandwidth, interconnects, and specialized compute.

For instance, the A100's memory clock is much lower than the RTX 4090's on paper (roughly 3 Gbps vs. 21 Gbps). However, the A100 uses **HBM2e memory** with a much wider **5,120-bit interface**. This design allows it to deliver around **2 TB/s** of bandwidth – double the RTX 4090 – despite the lower frequency. It's an approach that prioritizes efficiency and scale.

#### To recap, here are a few notable features of the A100:

-   **80 GB of HBM2e memory** with about **2 TB/s of bandwidth** (twice that of the RTX 4090).  

-   **5,120-bit memory bus**, an exceptionally wide interface enabling fast access to memory, reducing bottlenecks in AI and high-performance computing (HPC) workloads.  

-   **432 third-gen Tensor Cores**, with high-throughput **TF32** performance for AI and robust **FP64** support for HPC workloads.  

-   **Multi-Instance GPU (MIG)** support, so it can be partitioned into up to 7 isolated GPU instances – ideal for multi-tenant workloads and fine-grained resource allocation.  

-   **NVLink and NVSwitch** allow up to 16 A100 GPUs to be interconnected at up to **600 GB/s** of bi-directional bandwidth per GPU.

In short, the A100 delivers serious performance for large-scale AI and HPC workloads. But if you're comparing it head-to-head with the RTX 4090 on specs alone, here's how the two stack up.

### RTX 4090 vs. A100: Key Specs Compared

For reference, this table highlights some of the features and specs of the RTX 4090 and A100:

<table class="w-full border-collapse border border-gray-300">
    <thead>
        <tr class="bg-gray-100">
            <th class="border border-gray-300 px-4 py-2 text-left font-semibold">
                Specification
            </th>
            <th class="border border-gray-300 px-4 py-2 text-center font-semibold">
                RTX 4090
            </th>
            <th class="border border-gray-300 px-4 py-2 text-left font-semibold">
                A100
            </th>
        </tr>
    </thead>
    <tbody>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Architecture
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                Ada Lovelace
            </td>
            <td class="border border-gray-300 px-4 py-2">Ampere</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">VRAM</td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                24 GB GDDR6X
            </td>
            <td class="border border-gray-300 px-4 py-2">80 GB HBM2e</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Memory Clock
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                1313 MHz 21 Gbps
            </td>
            <td class="border border-gray-300 px-4 py-2">1512 MHz 3 Gbps</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Memory Bus Width
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">384</td>
            <td class="border border-gray-300 px-4 py-2">5120</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Bandwidth
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                1.01 TB/s
            </td>
            <td class="border border-gray-300 px-4 py-2">1.94 TB/s</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Streaming Multiprocessors (SMs)
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">128</td>
            <td class="border border-gray-300 px-4 py-2">108</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Tensor Cores
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">512</td>
            <td class="border border-gray-300 px-4 py-2">432</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                CUDA Cores
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">16,384</td>
            <td class="border border-gray-300 px-4 py-2">6,912</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Ray Tracing Cores
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">128</td>
            <td class="border border-gray-300 px-4 py-2">N/A</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Base Clock
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                2235 MHz
            </td>
            <td class="border border-gray-300 px-4 py-2">1065 MHz</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Boost Clock
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                2520 MHz
            </td>
            <td class="border border-gray-300 px-4 py-2">1410 MHz</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Multi-Instance GPUs
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                No MIG Support
            </td>
            <td class="border border-gray-300 px-4 py-2">
                Up to 7 MIGs @ 10GB each
            </td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Thermal Design Power (TDP)
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">450W</td>
            <td class="border border-gray-300 px-4 py-2">300W</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Recommended Power Supply
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">850W</td>
            <td class="border border-gray-300 px-4 py-2">700W</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Launch Date
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                Sept. 20, 2022
            </td>
            <td class="border border-gray-300 px-4 py-2">June 28, 2021</td>
        </tr>
    </tbody>
</table>

As mentioned, specs are just one part of the equation. Ultimately, choosing the right GPU depends on your workload, environment, and priorities.

### Use Cases: Which GPU Should You Choose?

Here’s how the strengths of each GPU align with different types of use cases.

#### You should consider the RTX 4090 if you prefer:

-   **A powerful, desktop-friendly GPU** for high-end gaming, creative workflows, and light AI experimentation.
-   **Excellent performance-per-dollar in single-GPU setups**, particularly for developers and small teams.
-   **A widely available, less costly option** for rendering, video editing, or running small to medium AI models locally.

#### Whereas the A100 may be a better fit if you need:

-   **Scalable performance** for training large models, running high-throughput inference, or powering complex simulations.
-   **A data center-optimized GPU** with advanced memory architecture and precision modes tailored for AI and HPC workloads.
-   **Flexible deployment options**, including support for multi-instance workloads and high-speed interconnects in clustered environments.

\------------------------------

At **Vast.ai**, we understand there are other considerations, as well. Access, flexibility, and affordability are just as important, especially when you're experimenting, scaling, or moving fast.

That's why our market-based cloud GPU rental platform gives you affordable, on-demand access to RTX 4090s, A100s, and much more. With Vast.ai, you can instantly match the right hardware to your workload – without the upfront investment and overhead of managing your own infrastructure.

Try Vast today and discover what you can achieve with compute power that fits your needs and your budget\!


Compare NVIDIA’s RTX 4090 vs A100 GPUs to find out which is better for gaming, AI, HPC, or creative work—and how to access both on Vast.ai.

NVIDIA RTX 4090 vs. A100: Two Powerhouses, Two Purposes


The H100 and H200 GPUs represent the top end of NVIDIA's current lineup for AI and high-performance computing (HPC). Both are designed for serious, large-scale workloads – from training foundation models to running complex scientific simulations.

At a glance, the two GPUs may appear similar, but for those working across enterprise environments or pushing the boundaries of exascale computing, the differences between them can have a major impact on performance and scalability.

Today, we're taking a look at how the H100 and H200 stack up against each other to help you decide which of these two heavyweights may be the better fit for your needs.

### NVIDIA H100: The Proven Performer

The H100 GPU made waves when it launched in March 2023\. Quickly adopted by big names like OpenAI, Meta, and Stability AI, it became the engine behind some of the most advanced generative AI systems in production today.

Built on NVIDIA's innovative Hopper architecture, the H100 was engineered for scale, handling everything from trillion-parameter language models to high-throughput inference. It introduced **fourth-gen Tensor Cores** and a dedicated **Transformer Engine**, designed to accelerate deep learning workloads with speed and FP8 precision.

The H100 boasts up to a **30X boost in performance** for large language models (LLMs) over its predecessor, the A100 GPU, according to NVIDIA. This massive improvement helped cement the H100 as the new standard for foundational model training, and to this day, it's still one of the very best GPUs on the market.

#### Some of the H100's more notable features include:

-   **80 GB of HBM3 memory**, with some models offering up to 94 GB.

-   **3.35 TB/s of memory bandwidth** to handle data-intensive workloads.

-   **Fourth-gen Tensor Cores** with superior throughput performance, along with an advanced Transformer Engine and FP64 and FP8 capabilities to reduce memory usage.

-   **PCIe Gen5 and fourth-gen NVLink** (with 900 GB/s of GPU-to-GPU interconnect), as well as NDR Quantum-2 InfiniBand networking for efficient scalability.

-   **Multi-Instance GPU (MIG) support** for secure partitioning into up to 7 isolated right-size instances that maximize quality of service for multi-tenant environments and parallel workloads.

If your projects involve massive datasets, large-scale simulations, or advanced AI training for foundational models, the H100 has proven itself as a reliable, high-performance solution. But with the arrival of the H200 in late 2024, NVIDIA has taken things a step further, introducing upgrades that push the envelope even more for memory-intensive and high-throughput workloads.

We ran a side-by-side comparison of inference performance on the **DeepSeek R1 Distill Llama 8B** model using `vLLM` with both NVIDIA H100 and H200 GPUs on Vast.ai.

#### **Benchmark Configuration**

-   **Model:** `deepseek-ai/DeepSeek-R1-Distill-Llama-8B`
-   **Requests:** 5,000
-   **Batch:** 512
-   **Input Tokens:** 16
-   **Output Tokens:** 128
-   **Backend:** `vLLM`

---

### **Results**

<table class="w-full border-collapse border border-gray-300">
    <thead>
        <tr class="bg-gray-100">
            <th class="border border-gray-300 px-4 py-2 text-left font-semibold">
                Metric
            </th>
            <th class="border border-gray-300 px-4 py-2 text-center font-semibold">
                H100
            </th>
            <th class="border border-gray-300 px-4 py-2 text-center font-semibold">
                H200
            </th>
        </tr>
    </thead>
    <tbody>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Benchmark Duration
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                45.28 s
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                40.02 s
            </td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Request Throughput
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                110.42 req/s
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                124.93 req/s
            </td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Output Throughput
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                13,650 tok/s
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                15,435 tok/s
            </td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Total Throughput
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                15,416 tok/s
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                17,434 tok/s
            </td>
        </tr>
    </tbody>
</table>

### NVIDIA H200: The Supercharged Successor

The H200 builds directly on the foundation laid by the H100, bringing meaningful improvements that target one of the biggest bottlenecks in AI and HPC workloads: memory.

While it shares the same Hopper architecture as the H100 and has similar core compute specs, the H200 significantly increases both memory capacity and bandwidth, two factors that become critical when operating at enterprise scale or pushing into exascale territory.

Equipped with **141 GB of HBM3e memory** and **4.8 TB/s of bandwidth**, the H200 nearly doubles the memory of the H100 and delivers about **1.4X faster** data access. Early benchmarks reflect that improvement. In MLPerf testing using Llama 2 70B, the H200 achieved over 31,000 tokens per second – about **45% faster than the H100**.

<img
    src="/uploads/h100vsh200_2.png"
    alt="Image credit: NVIDIA"
    style={{ width: "100%" }}
/>

The H200 retains all of the architectural strengths of the H100, including **fourth-gen NVLink**, **PCIe Gen5**, **MIG support**, and full **Tensor Core and Transformer Engine capabilities**. But for developers and organizations looking to squeeze out more performance without overhauling their infrastructure, the H200 offers a straightforward upgrade path with [measurable gains](https://resources.nvidia.com/en-us-data-center-overview/hpc-datasheet-sc23-h200).

#### To recap, here are a few advantages of the H200:

-   **141 GB of HBM3e memory at 4.8 TB/s** – nearly double the capacity of the H100, with 1.4X more memory bandwidth.

-   **Up to 1.9X the inference performance** of the H100 when handling LLMs, with more efficient processing per token and per inference cycle.

-   **50% reduced energy use and total cost of ownership** (TCO) due to improved thermal management and energy efficiency.

-   **Drop-in compatibility with existing Hopper-based systems**, enabling accelerated performance gains in H100-optimized infrastructure with minimal or no additional changes.

In short, the H200 represents a meaningful leap beyond the H100 in memory capacity, bandwidth, and throughput, bringing a new level of performance to AI and HPC workloads. That comes at a premium price, of course. But if you're weighing the tradeoffs between the two GPUs in terms of specs alone, let's take a side-by-side look at how they compare.

### H100 vs. H200: Key Specs Compared

For reference, here's a quick outline of the various features and specs of the NVIDIA H100 and H200:

<table class="w-full border-collapse border border-gray-300">
    <thead>
        <tr class="bg-gray-100">
            <th class="border border-gray-300 px-4 py-2 text-left font-semibold">
                Specification
            </th>
            <th class="border border-gray-300 px-4 py-2 text-center font-semibold">
                H100
            </th>
            <th class="border border-gray-300 px-4 py-2 text-center font-semibold">
                H200
            </th>
        </tr>
    </thead>
    <tbody>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                GPU Architecture
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">Hopper</td>
            <td class="border border-gray-300 px-4 py-2 text-center">Hopper</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                GPU Memory
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                80 GB HBM3
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                141 GB HBM3e
            </td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                GPU Memory Bandwidth
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                3.35 TB/s
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                4.8 TB/s
            </td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                FP64 Tensor Core
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                33.5 TFLOPS
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                33.5 TFLOPS
            </td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">FP32</td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                67 TFLOPS
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                67 TFLOPS
            </td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                TF32 Tensor Core
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                989 TFLOPS
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                989 TFLOPS
            </td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                FP16 Tensor Core
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                1,979 TFLOPS
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                1,979 TFLOPS
            </td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                FP8 Tensor Core
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                3,958 TFLOPS
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                3,958 TFLOPS
            </td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                INT8 Tensor Core
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                3,958 TOPS
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                3,958 TOPS
            </td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Decoders
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                7 NVDEC, 7 NVJPEG
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                7 NVDEC, 7 NVJPEG
            </td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                L2 Cache
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">50 MB</td>
            <td class="border border-gray-300 px-4 py-2 text-center">50 MB</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Max. Thermal Design Power
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                Up to 700 W
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                Up to 1000 W
            </td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Multi-Instance GPUs
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                Up to 7 MIGs @ 10GB each
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                Up to 7 MIGs @ 18GB each
            </td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Form Factor
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">SXM</td>
            <td class="border border-gray-300 px-4 py-2 text-center">SXM</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Interconnect
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                PCIe 5.0 x16
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                PCIe 5.0 x16
            </td>
        </tr>
    </tbody>
</table>

Ultimately, choosing the right GPU largely depends on your workload demands and performance priorities.

### Use Cases: Which GPU Should You Choose?

While both the H100 and H200 excel in AI and HPC environments, they serve slightly different needs.

#### You should consider the H100 if you prefer:

-   **A proven, high-performance GPU** for training medium to large models and handling intensive AI or HPC workloads.
-   **Use cases involving compute-bound workloads**, where total memory capacity is not as critical.
-   **A time-tested solution** already in widespread production use, balancing performance with lower power and cost.

#### Whereas the H200 may be a better fit if you need:

-   **Greater memory capacity and bandwidth** for running _even larger_ models, handling memory-bound workloads, or scaling LLM inference on the enterprise level.
-   **Higher throughput**, especially for tasks involving extended context windows, multi-modal AI, or very large batch sizes.
-   **A forward-looking upgrade path** with improved energy efficiency and lower TCO, thanks to higher performance per watt.

\------------------------------

At **Vast.ai**, we understand that choosing a GPU isn't only about specs. Often, the decision comes down to access and affordability, as well – especially when we're talking about some of the most powerful and expensive GPUs in existence. (Speaking of, check out our previous posts comparing the [H100 vs. A100](https://vast.ai/article/H100-vs-A100-Comparing-two-Powerhouse-GPUs), and the [H100 vs. L40S](https://vast.ai/article/nvidia-h100-vs-l40s-power-meets-versatility).)

That's why our cloud GPU rental platform is built to offer high-performance compute at a fraction of the typical cost. Vast gives you flexible, cost-effective access to H100s, H200s, and more.

With options for on-demand pricing and budget-friendly spot instances, you can scale up as needed without overspending. On average, **Vast users save 5- 6X on GPU compute** compared to traditional cloud providers.

Explore our platform today and get the power you need, when you need it – with no upfront investment required.


The H100 and H200 GPUs represent the top end of NVIDIA's current lineup for AI and high-performance computing (HPC). Both are designed for serious, large-scale workloads – from training foundation models to running complex scientific simulations.

NVIDIA H100 vs. H200: Two Hopper-based Heavyweights


When it comes to high-performance computing (HPC), the NVIDIA H100 is among the best of the best GPUs on the market. Its predecessor, the NVIDIA A100, is also a very impressive GPU in its own right. From enterprise to exascale, these GPUs are enabling groundbreaking advancements in everything from AI research to complex scientific simulations.

Today we're taking a look at how the H100 and A100 compare with each other. Depending on your use case, one or the other might be better suited for your needs and preferences. So let's dive in!

## Introduction to the NVIDIA H100 and A100

As mentioned, the H100 GPU is the successor to the A100. It was specifically optimized for machine learning and designed to [deliver](https://developer.nvidia.com/blog/nvidia-hopper-architecture-in-depth/) an "order-of-magnitude performance leap for large-scale AI and HPC" over the A100, with "substantial improvements in architectural efficiency."

Based on the innovative Hopper architecture, the H100 can speed up large language models (LLMs) by 30X over the A100, according to NVIDIA. It features fourth-gen Tensor Cores and – unlike the A100 – a dedicated Transformer Engine to solve trillion-parameter language models.

Whether it's used in smaller enterprise systems or massive, unified GPU clusters, the H100 delivers efficient scalability thanks to its combination of fourth-gen NVLink (with 900 GB/s of GPU-to-GPU interconnect), NDR Quantum-2 InfiniBand networking (speeding up communication by every GPU across nodes), and PCIe Gen5. With superior FP64 and FP8 performance, the H100 is ideal for next-gen AI workloads.

The A100 GPU is known for its versatility, however. Built on the Ampere architecture, it was groundbreaking when released in 2020, providing up to 20X higher performance over the prior NVIDIA Volta generation. It can efficiently scale up or be partitioned using Multi-Instance GPU (MIG) into seven isolated GPU instances, just like the H100, which is ideal for shifting workload demands.

In short, the A100 excels in both training and inference tasks, and while it may not offer the sheer power of the H100, its robust FP64 performance makes it a great fit for scientific simulations and data analytics.

## Key Differences: H100 vs. A100

The H100 has an updated chip design and various other factors that distinguish it from the A100. The following are a few that are important to consider:

### Memory Bandwidth

The A100 has a memory bandwidth of about 2 TB/s, which is plenty for most use cases; however, the H100 offers an impressive 3.35 TB/s, making it ideal for workloads where data transfer bottlenecks may occur.

### CUDA Cores and Tensor Cores

A higher number of CUDA and Tensor Cores means a GPU can perform more parallel computations simultaneously. With 14,592 CUDA Cores and 456 Tensor Cores, the H100 outperforms the A100’s 6,912 CUDA Cores and 432 Tensor Cores, delivering superior performance in applications that can leverage increased parallelism.

### Confidential Computing

The H100's Hopper architecture introduced NVIDIA's new confidential computing environment. It protects data-in-use by performing computation in a trusted execution environment (TEE) that is both hardware-based and attested. This means you essentially have absolute security for use cases where security, privacy, and regulatory compliance are paramount.

### Tensor Memory Accelerator (TMA)

Another feature of the H100's Hopper architecture is the TMA unit, which can transfer large blocks of data efficiently between global memory and shared memory. It supports asynchronous copies between thread blocks in a cluster as well, optimizing data movement and memory management for tensor operations. Ultimately, this enhances the efficiency and speed of large-scale AI and deep learning tasks.

### MIG Capability

The Multi-Instance GPU (MIG) feature is present in both the H100 and A100, providing flexible workload management so you can run multiple LLMs in parallel, for instance. The H100's MIG capability is more robust, however, with better resource allocation and more memory and compute power to split.

## Performance Benchmarks

According to NVIDIA's own benchmarks, the H100 offers up to 30X better inference performance than the A100. In another test where the A100 and H100 were used in clusters to train two large language models (LLMs), the H100 demonstrated remarkable speed improvements. The H100's new NVLink Switch System provided an even bigger boost, enabling the H100 cluster to train the models up to nine times faster than the A100 cluster:

![H100 vs A100 Speed Comparison](/uploads/Speed-up-over-A100.png)

## Choosing the Right GPU

So what does all this mean for you? The best GPU for your needs will depend on numerous factors. Here's a quick outline of the various features and specs of the NVIDIA A100 and H100:

<table class="w-full border-collapse border border-gray-300">
  <thead>
    <tr class="bg-gray-100">
      <th class="border border-gray-300 px-4 py-2 text-left font-semibold">Feature</th>
      <th class="border border-gray-300 px-4 py-2 text-center font-semibold">A100</th>
      <th class="border border-gray-300 px-4 py-2 text-center font-semibold">H100</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td class="border border-gray-300 px-4 py-2 font-semibold">GPU Architecture</td>
      <td class="border border-gray-300 px-4 py-2 text-center">Ampere</td>
      <td class="border border-gray-300 px-4 py-2 text-center">Hopper</td>
    </tr>
    <tr class="bg-gray-50">
      <td class="border border-gray-300 px-4 py-2 font-semibold">GPU Memory</td>
      <td class="border border-gray-300 px-4 py-2 text-center">40 or 80 GB HBM2e</td>
      <td class="border border-gray-300 px-4 py-2 text-center">80 MB HBM3</td>
    </tr>
    <tr>
      <td class="border border-gray-300 px-4 py-2 font-semibold">GPU Memory Bandwidth</td>
      <td class="border border-gray-300 px-4 py-2 text-center">1.6 to 2 TB/s</td>
      <td class="border border-gray-300 px-4 py-2 text-center">3.35 TB/s</td>
    </tr>
    <tr class="bg-gray-50">
      <td class="border border-gray-300 px-4 py-2 font-semibold">CUDA Cores</td>
      <td class="border border-gray-300 px-4 py-2 text-center">6912</td>
      <td class="border border-gray-300 px-4 py-2 text-center">14,592</td>
    </tr>
    <tr>
      <td class="border border-gray-300 px-4 py-2 font-semibold">FP64 TFLOPS</td>
      <td class="border border-gray-300 px-4 py-2 text-center">9.7</td>
      <td class="border border-gray-300 px-4 py-2 text-center">33.5</td>
    </tr>
    <tr class="bg-gray-50">
      <td class="border border-gray-300 px-4 py-2 font-semibold">FP32 TFLOPS</td>
      <td class="border border-gray-300 px-4 py-2 text-center">19.6</td>
      <td class="border border-gray-300 px-4 py-2 text-center">67</td>
    </tr>
    <tr>
      <td class="border border-gray-300 px-4 py-2 font-semibold">TF32 Tensor Core Flops*</td>
      <td class="border border-gray-300 px-4 py-2 text-center">156 | 312</td>
      <td class="border border-gray-300 px-4 py-2 text-center">378 | 756</td>
    </tr>
    <tr class="bg-gray-50">
      <td class="border border-gray-300 px-4 py-2 font-semibold">FP16 Tensor Core Flops*</td>
      <td class="border border-gray-300 px-4 py-2 text-center">312 | 624</td>
      <td class="border border-gray-300 px-4 py-2 text-center">756 | 1513</td>
    </tr>
    <tr>
      <td class="border border-gray-300 px-4 py-2 font-semibold">FP8 Tensor Core TFLOPS</td>
      <td class="border border-gray-300 px-4 py-2 text-center">N/A</td>
      <td class="border border-gray-300 px-4 py-2 text-center">3958</td>
    </tr>
    <tr class="bg-gray-50">
      <td class="border border-gray-300 px-4 py-2 font-semibold">Peak INT8 TOPS*</td>
      <td class="border border-gray-300 px-4 py-2 text-center">624 | 1248</td>
      <td class="border border-gray-300 px-4 py-2 text-center">3958</td>
    </tr>
    <tr>
      <td class="border border-gray-300 px-4 py-2 font-semibold">Media Engine</td>
      <td class="border border-gray-300 px-4 py-2 text-center">0 NVENC<br>5 NVDEC<br>5 NVJPEG</td>
      <td class="border border-gray-300 px-4 py-2 text-center">0 NVENC<br>7 NVDEC<br>7 NVJPEG</td>
    </tr>
    <tr class="bg-gray-50">
      <td class="border border-gray-300 px-4 py-2 font-semibold">L2 Cache</td>
      <td class="border border-gray-300 px-4 py-2 text-center">40 MB</td>
      <td class="border border-gray-300 px-4 py-2 text-center">50 MB</td>
    </tr>
    <tr>
      <td class="border border-gray-300 px-4 py-2 font-semibold">Power</td>
      <td class="border border-gray-300 px-4 py-2 text-center">Up to 400 W</td>
      <td class="border border-gray-300 px-4 py-2 text-center">Up to 700 W</td>
    </tr>
    <tr class="bg-gray-50">
      <td class="border border-gray-300 px-4 py-2 font-semibold">Form Factor</td>
      <td class="border border-gray-300 px-4 py-2 text-center">SXM4 - 8-way HGX</td>
      <td class="border border-gray-300 px-4 py-2 text-center">SXM5 - 8-way HGX</td>
    </tr>
    <tr>
      <td class="border border-gray-300 px-4 py-2 font-semibold">Interconnect</td>
      <td class="border border-gray-300 px-4 py-2 text-center">PCIe 4.0 x16</td>
      <td class="border border-gray-300 px-4 py-2 text-center">PCIe 5.0 x16</td>
    </tr>
  </tbody>
</table>

\*Without and with structured sparsity.

---

## Use Cases

When it comes to use cases, you may want to consider some general guidelines.

### The A100 is well suited for:

-   Versatile AI workloads including training and inference.
-   Applications needing a balance of performance and cost-efficiency.
-   Scenarios where ultra-high precision isn't the priority.

### The H100 is ideal for:

-   Cutting-edge AI training, large-scale simulations, and other HPC workloads.
-   Environments where computational speed and memory bandwidth are critical.
-   Future-proofing investments in AI infrastructure.

---

When deciding between the A100 and H100, consider your specific workload requirements. If you need top-tier double-precision performance and superior memory bandwidth, or you're dealing with next-gen HPC at datacenter scale and trillion-parameter AI, the H100 is the clear winner. For a more versatile and cost-effective solution that still delivers powerful AI performance, the A100 is a solid choice.

At Vast.ai, we understand the importance of finding the right GPU for your needs as well as your budget. Our cloud GPU rental marketplace offers a variety of machines at the lowest prices possible, saving you 5-6X on GPU compute. Whether you prefer on-demand pricing or interruptible instances through spot auction-based pricing, we've got you covered.

We invite you to discover what Vast can do for you and accelerate your projects with the right GPU today!


Explore the differences between NVIDIA's H100 and A100 GPUs, from architectural advancements to performance benchmarks.

H100 vs A100: Comparing Two Powerhouse GPUs


With all of the GPU options, finding the one that's right for your needs is challenging. Whether you're into high-end cloud gaming, training AI models, or even outfitting a university lab or data center, understanding the nuances between different models is key.

Today, we’re focusing on two powerful GPUs from NVIDIA: the L40 and [L40S](https://cloud.vast.ai/?gpu_option=L40S). While they may not have received as much attention as some of the company's other offerings, they certainly deserve a closer look! These two GPUs, built on the [Ada Lovelace](https://vast.ai/article/rtx-4090-deep-learning) architecture, pack a punch in performance and capabilities.

But how do they stack up against each other? What about compared to other GPUs? And more importantly, which one should you choose for your specific project needs?

## The Basics: NVIDIA L40 vs. L40S

The L40S is essentially an upgrade to the L40 – as well as a close relative to the [RTX 4090 gaming graphics card](https://vast.ai/article/nvidia-rtx-4090-vs-5090).

Initially designed for data center graphics and simulation tasks, the L40 is the engine of NVIDIA Omniverse. It's a powerhouse for extended (XR) and virtual reality (VR) applications, design collaboration, and digital twins, thanks to its advanced RTX and AI capabilities. It enables accelerated ray-traced and path-traced rendering, delivers physically accurate simulations, and generates highly realistic 3D synthetic data for the most demanding Omniverse workloads.

Compared to the previous generation, the L40 GPU delivers 5X higher inference performance for compute-intensive AI workloads – so it can generate high-quality images and immersive visual content lightning fast. It's perfectly suited for image generative AI applications. If that's all you need, then the L40 might be an ideal option for you.

However, the L40S can do all of this and more. It was developed to meet the surging demand for GPUs that can handle the intense computational requirements of machine learning training and inference.

NVIDIA describes the L40S as "the most powerful universal GPU," offering "breakthrough multi-workload performance." Here's a quick snapshot of its performance in a couple of areas:

!["Graph comparing NVIDIA L40S GPU performance for Stable Diffusion image generation and LLM inference, showing significant differences in images per minute and token latency across various configurations."](/uploads/l4-vs-l40S-image-generative-ai.png)

After its release in late 2023, the L40S achieved popularity due to the lack of availability of the NVIDIA A100 and H100 Tensor Core GPUs. There was a massive chip shortage and backlogged shipments, forcing many buyers to turn to other options – like the L40S. Since the L40S can not only be used for the same workloads as the L40 but can also power AI training and inference at a high level like the A100 and H100, it was a logical alternative.

The L40S, A100, and H100 GPUs are therefore in a unique position to be compared with each other. Let's take a look!

## L40S vs. A100 vs. H100: Specs and Performance

Each of the three GPUs brings its own strengths to the table. The table below compares various features and specs between them.

| Feature                | A100          | L40S           | H100          |
|------------------------|---------------|----------------|---------------|
| **GPU Architecture**   | Ampere        | Ada Lovelace   | Hopper        |
| **GPU Memory**         | 40 or 80 GB HBM2e | 48 GB GDDR6    | 80 MB HBM3    |
| **GPU Memory Bandwidth** | 1.6 to 2 TB/s | 864 GB/s       | 3.35 TB/s     |
| **CUDA Cores**         | 6912          | 18,176         | 14,592        |
| **FP64 TFLOPS**        | 9.7           | N/A            | 33.5          |
| **FP32 TFLOPS**        | 19.6          | 91.6           | 67            |
| **TF32 Tensor Core Flops** | 156 | 312 | 183 | 366 | 378 | 756 |
| **FP16 Tensor Core Flops** | 312 | 624 | 362 | 733 | 756 | 1513 |
| **FP8 Tensor Core TFLOPS** | N/A         | 1466           | 3958          |
| **Peak INT8 TOPS**     | 624 | 1248   | 733 | 1466      | 3958          |
| **Media Engine**       | 0 NVENC 5 NVDEC 5 NVJPEG | 0 NVENC 5 NVDEC 5 NVJPEG | 0 NVENC 7 NVDEC 7 NVJPEG |
| **L2 Cache**           | 40 MB         | 96 MB          | 50 MB         |
| **Power**              | Up to 400 W   | Up to 350 W    | Up to 700 W   |
| **Form Factor**        | SXM4 - 8-way HGX | Dual Slot Width | SXM5 - 8-way HGX |
| **Interconnect**       | PCIe 4.0 x16  | PCIe 4.0 x16   | PCIe 5.0 x16  |

*Without and with structured sparsity.

As shown, there are some clear differences in performance when it comes to FP64 (double-precision), FP32 (single-precision), and FP16 (half-precision) computations.

For instance, the L40S does not natively support FP64. This means that it may not perform as well as the A100 and H100 in applications that require high precision. The H100 in particular, with its notably superior FP64 performance, is generally preferred for the most demanding tasks.

In terms of FP32 and FP16 Tensor Core performance, the L40S outperforms the A100 40GB and can hold its own against the H100. However, in memory-intensive machine learning scenarios, the L40S's lower memory bandwidth may counterbalance its performance compared to both the A100 80GB and H100.

Still, the L40S can really shine next to the A100 in certain situations. According to NVIDIA, when it comes to complex AI workloads involving billions of parameters and multiple data modalities, including text and video, the L40S can [achieve](https://nvidianews.nvidia.com/news/nvidia-global-data-center-system-manufacturers-to-supercharge-generative-ai-and-industrial-digitalization) up to 1.2 times the generative AI inference performance and up to 1.7 times the training performance of the A100.

Another point in the L40S's favor? While it may not be able to beat the H100 on certain metrics, it's definitely more flexible. The higher-performance H100 can't be used as an Omniverse server and doesn't support graphics, whereas the L40S can be and does.

## The Bottom Line


The NVIDIA L40 excels in powering image generative AI applications and Omniverse workloads. You may not need it for much else, and if that's the case, the L40 would probably serve you well.

On the other hand, the NVIDIA L40S nicely balances versatility as well as performance. The following are some reasons you might choose it for your next project:

- **You have multi-modal workloads** – You'll achieve remarkable efficiency with the L40S, enabling you to run HPC simulations, train AI models, and render images using the same computing infrastructure. Plus, video output is built in.
- **You need compute power fast in a familiar form factor** – The L40S is designed for rapid deployment within existing systems, offering powerful performance without requiring extensive reconfiguration. Its Dual Slot PCIe form factor ensures compatibility with a wide range of setups.
- **Your workloads don't require extreme precision or maximum speed** – Without the math performance (FLOPS), high-bandwidth memory, and NVLINK of the H100, the L40S's inference and training performance for smaller models is nonetheless perfectly suitable in most cases.
- **You prioritize cost-effectiveness** – The L40S provides an excellent price-to-performance ratio, particularly for enterprises and research institutions that need robust AI capabilities without breaking the budget.

For many of us, cost is a major factor in [GPU decision-making](https://vast.ai/pricing). At Vast.ai, our mission is to democratize AI and bring massive compute power, at an affordable price, to anyone who needs it. You don't have to make a huge hardware investment upfront.

Our cloud GPU rental marketplace offers the lowest prices possible, enabling you to access top-tier machines at any time – and save 5-6X on GPU compute. We offer low-cost on-demand pricing as well as interruptible instances through spot auction-based pricing to bring you even more savings.

Come explore what Vast can do for you and accelerate your projects today!

Comparing NVIDIA L40 vs. L40s – and More


For both technology enthusiasts and industry professionals, keeping a close eye on the latest GPU advancements isn't just a hobby – it's crucial for staying competitive and maximizing performance and compute power. Here at Vast.ai, we're just as excited to track the biggest leaps forward in graphics technology.

Today, we're looking at two GPUs in particular: the **NVIDIA RTX 4090** and its formidable successor, the **RTX 5090**. The RTX 4090 has already earned its reputation for performance, while the RTX 5090 has recently made its debut and is setting new standards in the industry.

Let's explore how the two compare with each other\!

### Spotlight on NVIDIA GeForce RTX 4090

The **RTX 4090** made waves as the top-tier consumer-grade GPU of its generation. Built on the AD102 chip and Ada Lovelace architecture, it quickly became the go-to choice for high-end gaming and demanding professional workloads.

With a triple-slot design, the RTX 4090 boasts **128 third-gen RT cores** and **512 fourth-gen Tensor cores**, elevating ray-tracing performance and speeding up AI-driven tasks like **DLSS 3** in games.

Plus, it's packed with **16,384 CUDA cores** and has **24GB of GDDR6X memory** with a maximum bandwidth of **1,008 GB/s** and a **384-bit memory bus** interface. Error-correcting code (ECC) can be enabled, but this reduces performance while making no difference in GPU rendering – so keeping it turned off is appropriate for most uses when you're prioritizing speed.

Due to its high processing capabilities, the RTX 4090 draws plenty of power: **450 watts**. It requires an 850-watt power supply and robust cooling solutions, making it one of the more demanding GPUs to run, to this day.

When it launched in October 2022, the RTX 4090 raised the bar for consumer GPUs. It may no longer sit at the top, but its capabilities continue to make it a powerhouse for serious workloads and high-end gaming.

### The New NVIDIA GeForce RTX 5090

The RTX 5090, released earlier this year, is a game-changer, redefining high-end performance in the consumer space with its groundbreaking features and next-gen architecture. With the RTX 50 series, we've now moved beyond NVIDIA's Ada Lovelace and into the era of the Blackwell architecture.

Joining the new Blackwell GPU lineup, the flagship GeForce RTX 5090 utilizes the "physically monolithic" **GB202 die**. Among its highlights are a **512-bit memory bus** and **32GB of next-gen GDDR7 memory** (the successor to the GDDR6/X memory used by the RTX 40-series GPUs). The RTX 5090 also boasts memory bandwidth of 1,792 GB/s, a substantial jump from the RTX 4090's 1,008 GB/s​.

Here's a quick side-by-side comparison of the RTX 5090 and RTX 4090's specs:

<table class="w-full border-collapse border border-gray-300">
    <thead>
        <tr class="bg-gray-100">
            <th class="border border-gray-300 px-4 py-2 text-left font-semibold">
                Spec
            </th>
            <th class="border border-gray-300 px-4 py-2 text-center font-semibold">
                RTX 4090
            </th>
            <th class="border border-gray-300 px-4 py-2 text-center font-semibold">
                RTX 5090
            </th>
        </tr>
    </thead>
    <tbody>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Architecture
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">AD102</td>
            <td class="border border-gray-300 px-4 py-2 text-center">GB202</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">VRAM</td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                24GB GDDR6X
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                32GB GDDR7
            </td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                VRAM Speed
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">21Gbps</td>
            <td class="border border-gray-300 px-4 py-2 text-center">28Gbps</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                VRAM Bus Width
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">384</td>
            <td class="border border-gray-300 px-4 py-2 text-center">512</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Bandwidth
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                1.01TB/s
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                1792GB/s
            </td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Streaming Multiprocessors (SMs)
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">128</td>
            <td class="border border-gray-300 px-4 py-2 text-center">170</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Tensor Cores
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">512</td>
            <td class="border border-gray-300 px-4 py-2 text-center">680</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Ray Tracing Cores
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">128</td>
            <td class="border border-gray-300 px-4 py-2 text-center">170</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Boost Clock
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                2520 MHz
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                2407 MHz
            </td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Total Graphics Power (TGP)
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">450W</td>
            <td class="border border-gray-300 px-4 py-2 text-center">575W</td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Recommended Power Supply
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">850W</td>
            <td class="border border-gray-300 px-4 py-2 text-center">1000W</td>
        </tr>
        <tr class="bg-gray-50">
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Launch Date
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                Sept. 20, 2022
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">
                Jan. 30, 2025
            </td>
        </tr>
        <tr>
            <td class="border border-gray-300 px-4 py-2 font-semibold">
                Launch Price
            </td>
            <td class="border border-gray-300 px-4 py-2 text-center">$1,599</td>
            <td class="border border-gray-300 px-4 py-2 text-center">$1,999</td>
        </tr>
    </tbody>
</table>

Additionally, the RTX 5090 comes equipped with **170 streaming multiprocessors** (SMs) and **21,760 CUDA cores**, as well as **170 fourth-gen RT cores** and **680 fifth-gen Tensor cores** – up from 128 and 512 respectively in the RTX 4090\.

In terms of power consumption, the RTX 5090 has a total graphics power (TGP) of **575 watts**, considerably higher than the RTX 4090's 450 watts.

When it comes to demanding AI and video workloads, the RTX 5090 shows clear gains over its predecessor. In Procyon's AI XL (FP16) benchmark, it outperformed the RTX 4090 by 40%, and in PugetBench's DaVinci Resolve test, it delivered a 12% performance boost.

Next-generation gaming features are a major draw for the RTX 5090\. One of the most significant leaps forward comes from **DLSS 4**, NVIDIA's AI-powered upscaling technology that further enhances performance and image quality. With advancements like **Multi Frame Generation**, **Neural Rendering**, and **Mega Geometry** – and with 4K 240FPS gaming becoming more attainable – the RTX 5090 is perfectly positioned to tackle the most demanding gaming scenarios and deliver the best possible gameplay experience.

(For a detailed examination of the RTX 5090, delving into its gaming features, real-world performance, and more, see our in-depth post [here](https://vast.ai/article/nvidia-geforce-rtx-5090-specs-everything-you-need-to-know).)

### Concluding Thoughts

Even after the release of the RTX 5090, it's important to recognize that the RTX 4090 still delivers exceptional performance, handling everything available at 4K with ease. The RTX 5090, while impressive, may be considered more of a luxury upgrade than a necessity for gamers and professionals who can (or already do) benefit from the power of the RTX 4090\. It really depends on whether the extra power will make enough of a difference for your needs.

Cost is another critical factor. The RTX 5090 is a top-of-the-line machine with a matching price tag. And the RTX 4090 is no slouch, either, still commanding a steep price to this day, over two years after its release.

Here at Vast.ai, we take pride in providing access to powerful machines for rent on the cloud at unbeatable prices. Vast offers unmatched flexibility, enabling you to utilize state-of-the-art technology without the expense and upkeep of purchasing your own hardware.

Whether you're looking to explore the capabilities of high-end consumer GPUs like the RTX 4090 or 5090, or check out any other machine on Vast, we'd be happy to help you discover the possibilities\!


A detailed comparison between NVIDIA's RTX 4090 and RTX 5090 GPUs, analyzing their performance, features, and value for both gaming and professional workloads.

Posts about: Technology

NVIDIA H200 vs. B200: Comparing Datacenter-Grade Accelerators

NVIDIA RTX 4090 vs. A100: Two Powerhouses, Two Purposes

NVIDIA H100 vs. H200: Two Hopper-based Heavyweights

H100 vs A100: Comparing Two Powerhouse GPUs

Comparing NVIDIA L40 vs. L40s – and More

NVIDIA RTX 4090 vs. RTX 5090: How They Stack Up