NVIDIA Jetson AGX Orin 32 GB
NVIDIA graphics card specifications and benchmark scores
Tensor CoresAt a Glance
NVIDIANVIDIA Jetson AGX Orin 32 GB Specifications
GPU Core
Shader units and compute resources
The NVIDIA Jetson AGX Orin 32 GB GPU core specifications define its raw processing power for graphics and compute workloads. Shading units (also called CUDA cores, stream processors, or execution units depending on manufacturer) handle the parallel calculations required for rendering. TMUs (Texture Mapping Units) process texture data, while ROPs (Render Output Units) handle final pixel output. Higher shader counts generally translate to better GPU benchmark performance, especially in demanding games and 3D applications.
Jetson AGX Orin 32 GB Clock Speeds
GPU and memory frequencies
Clock speeds directly impact the Jetson AGX Orin 32 GB's performance in GPU benchmarks and real-world gaming. The base clock represents the minimum guaranteed frequency, while the boost clock indicates peak performance under optimal thermal conditions. Memory clock speed affects texture loading and frame buffer operations. The Jetson AGX Orin 32 GB by NVIDIA dynamically adjusts frequencies based on workload, temperature, and power limits to maximize performance while maintaining stability.
NVIDIA's Jetson AGX Orin 32 GB Memory
VRAM capacity and bandwidth
VRAM (Video RAM) is dedicated memory for storing textures, frame buffers, and shader data. The Jetson AGX Orin 32 GB's memory capacity determines how well it handles high-resolution textures and multiple displays. Memory bandwidth, measured in GB/s, affects how quickly data moves between the GPU and VRAM. Higher bandwidth improves performance in memory-intensive scenarios like 4K gaming. The memory bus width and type (GDDR6, GDDR6X, HBM) significantly influence overall GPU benchmark scores.
Jetson AGX Orin 32 GB by NVIDIA Cache
On-chip cache hierarchy
On-chip cache provides ultra-fast data access for the Jetson AGX Orin 32 GB, reducing the need to fetch data from slower VRAM. L1 and L2 caches store frequently accessed data close to the compute units. AMD's Infinity Cache (L3) dramatically increases effective bandwidth, improving GPU benchmark performance without requiring wider memory buses. Larger cache sizes help maintain high frame rates in memory-bound scenarios and reduce power consumption by minimizing VRAM accesses.
Jetson AGX Orin 32 GB Theoretical Performance
Compute and fill rates
Theoretical performance metrics provide a baseline for comparing the NVIDIA Jetson AGX Orin 32 GB against other graphics cards. FP32 (single-precision) performance, measured in TFLOPS, indicates compute capability for gaming and general GPU workloads. FP64 (double-precision) matters for scientific computing. Pixel and texture fill rates determine how quickly the GPU can render complex scenes. While real-world GPU benchmark results depend on many factors, these specifications help predict relative performance levels.
Jetson AGX Orin 32 GB Ray Tracing & AI
Hardware acceleration features
The NVIDIA Jetson AGX Orin 32 GB includes dedicated hardware for ray tracing and AI acceleration. RT cores handle real-time ray tracing calculations for realistic lighting, reflections, and shadows in supported games. Tensor cores (NVIDIA) or XMX cores (Intel) accelerate AI workloads including DLSS, FSR, and XeSS upscaling technologies. These features enable higher visual quality without proportional performance costs, making the Jetson AGX Orin 32 GB capable of delivering both stunning graphics and smooth frame rates in modern titles.
Ampere Architecture & Process
Manufacturing and design details
The NVIDIA Jetson AGX Orin 32 GB is built on NVIDIA's Ampere architecture, which defines how the GPU processes graphics and compute workloads. The manufacturing process node affects power efficiency, thermal characteristics, and maximum clock speeds. Smaller process nodes pack more transistors into the same die area, enabling higher performance per watt. Understanding the architecture helps predict how the Jetson AGX Orin 32 GB will perform in GPU benchmarks compared to previous generations.
Power & Thermal
TDP and power requirements
Power specifications for the NVIDIA Jetson AGX Orin 32 GB determine PSU requirements and thermal management needs. TDP (Thermal Design Power) indicates the heat output under typical loads, guiding cooler selection. Power connector requirements ensure adequate power delivery for stable operation during demanding GPU benchmarks. The suggested PSU wattage accounts for the entire system, not just the graphics card. Efficient power delivery enables the Jetson AGX Orin 32 GB to maintain boost clocks without throttling.
Jetson AGX Orin 32 GB by NVIDIA Physical & Connectivity
Dimensions and outputs
Physical dimensions of the NVIDIA Jetson AGX Orin 32 GB are critical for case compatibility. Card length, height, and slot width determine whether it fits in your chassis. The PCIe interface version affects bandwidth for communication with the CPU. Display outputs define monitor connectivity options, with modern cards supporting multiple high-resolution displays simultaneously. Verify these specifications against your case and motherboard before purchasing to ensure a proper fit.
NVIDIA API Support
Graphics and compute APIs
API support determines which games and applications can fully utilize the NVIDIA Jetson AGX Orin 32 GB. DirectX 12 Ultimate enables advanced features like ray tracing and variable rate shading. Vulkan provides cross-platform graphics capabilities with low-level hardware access. OpenGL remains important for professional applications and older games. CUDA (NVIDIA) and OpenCL enable GPU compute for video editing, 3D rendering, and scientific applications. Higher API versions unlock newer graphical features in GPU benchmarks and games.
Jetson AGX Orin 32 GB Product Information
Release and pricing details
The NVIDIA Jetson AGX Orin 32 GB is manufactured by NVIDIA as part of their graphics card lineup. Release date and launch pricing provide context for comparing GPU benchmark results with competing products from the same era. Understanding the product lifecycle helps evaluate whether the Jetson AGX Orin 32 GB by NVIDIA represents good value at current market prices. Predecessor and successor information aids in tracking generational improvements and planning future upgrades.
About NVIDIA Jetson AGX Orin 32 GB
The NVIDIA Jetson AGX Orin 32 GB is an Ampere-generation integrated processor from NVIDIA, built on Samsung's 8 nm process. The GA10B die measures 200 mm² and contains 1,792 shading units, 56 texture mapping units, 24 ROPs, and 56 tensor cores. It is a Tegra (Ampere) part with a 40 W TDP, an IGP slot width, and a PCIe 4.0 x4 bus interface. The production status is end-of-life, and the listed launch MSRP is 899 USD.
Memory Subsystem
The memory subsystem is defined by a 32 GB LPDDR5 pool. That capacity is substantial for an integrated device, and it is paired with a 256-bit memory bus. The memory clock is listed as 800 MHz with a 6.4 Gbps effective data rate, which produces a memory bandwidth of 204.8 GB/s. These figures are directly related: the effective data rate, the bus width, and the resulting bandwidth are all specified in the same record, giving a clear view of the memory path.
For high-resolution workloads, the large capacity matters because it determines how much geometry, texture data, and render-target state can remain resident. A 32 GB pool is large enough to address projects that would exceed the capacity of smaller embedded parts. But capacity is not the only constraint. The 204.8 GB/s bandwidth sets the sustained data-transfer limit between memory and the compute units. High resolutions increase both footprint and traffic; the bus must feed the 1,792 shading units, the 56 TMUs, and the 24 ROPs at the rates they are designed to consume.
The LPDDR5 type is important in an integrated context. The part is not a discrete Graphik board; its slot width is IGP, and display outputs are portable-device dependent. That means the memory interface is integrated into the same package and is used by the compute block rather than through a standalone card connector. The 256-bit bus width is a wide path for an integrated part, and the 6.4 Gbps effective signaling is the rate that turns that width into the 204.8 GB/s figure. In memory-bound scenarios, the bandwidth is the limiting factor, not the 32 GB capacity.
Ray Tracing and Feature Set
The Fact Pack does not list an RT core count for this device. The rtCores field is null, so no ray tracing hardware quantity is recorded. That absence is a meaningful qualifier: it is not a confirmation that ray tracing hardware exists, nor is it a claim that it does not. The available feature-level evidence comes from the API list. The device is recorded with DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.
DirectX 12 Ultimate (12_2) is the API tier listed in the Fact Pack. The exact feature content of that tier is not expanded in the data, so the record itself does not itemize individual ray tracing or mesh-shading capabilities. What is present is a clear API compatibility statement. Vulkan 1.4 and OpenGL 4.6 are also present, which gives a broad API surface for portable-device workloads.
The tensor core count is 56. These tensor cores sit alongside the 1,792 shading units and are a distinguishing part of the Ampere architecture. The Fact Pack does not provide separate tensor-core throughput values. It does list FP16 compute at 6.666 TFLOPS with a 2:1 ratio, and FP32 at 3.333 TFLOPS, but that ratio is described as a general compute ratio rather than a tensor-core-specific figure. For ray tracing, the only specification-level data is the missing RT core count and the API list.
How It Compares
The nearestRivals array in the Fact Pack is empty. There are no rival names, no rival scores, and no deltaPct values to report. As a result, direct comparative statements against named competing products cannot be grounded in the provided data.
The only relative ranking marker is percentileVsAllGpus, which is 50. This places the NVIDIA Jetson AGX Orin 32 GB at the midpoint of the all-GPU distribution tracked by the database. The avgBenchmarkScore field is 0, and the benchmarks array is empty, so this midpoint position is not backed by a nonzero averaged score. It is a positional indicator, not a measured performance figure.
Physically, this is a compact integrated part. It has a length of 100 mm (3.9 inches) and a height of 87 mm (3.4 inches). The bus interface is PCIe 4.0 x4, and the slot width is IGP. No power connector is listed, which is consistent with an integrated design rather than an add-in board. The production status is end-of-life.
Because there are no nearest rivals, this section cannot provide a per-rival paragraph with exact deltas. The comparison that can be made is against the database distribution: at percentile 50, the device sits at the median. The empty benchmark list means that any future comparison would need to be populated with measured scores before a percentage delta could be computed.
FAQ
Q: How much memory does the NVIDIA Jetson AGX Orin 32 GB have, and what type is it?
A: It has 32 GB of LPDDR5 memory on a 256-bit bus. The memory clock is 800 MHz, the effective data rate is 6.4 Gbps, and the resulting bandwidth is 204.8 GB/s.
Q: Does the Fact Pack list ray tracing cores for this device?
A: No. The rtCores field is null. The API list does include DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, but no RT core count is recorded.
Q: How many tensor cores are present?
A: The device contains 56 tensor cores. It also has 1,792 shading units, 56 texture mapping units, and 24 ROPs.
Q: What are the physical dimensions and form factor?
A: The length is 100 mm (3.9 inches) and the height is 87 mm (3.4 inches). The slot width is IGP, the bus interface is PCIe 4.0 x4, and display outputs are portable device dependent.
Q: What process node and die information is recorded?
A: The chip is GA10B, built on Samsung's 8 nm process, with a die size of 200 mm².
Q: What is the production status?
A: The production status is end-of-life.
Benchmark Performance
The benchmark records for this entry are empty. The benchmarks array contains no score rows, and the avgBenchmarkScore is 0. Because the nearestRivals array is also empty, there are no exact percentage deltas to present against named rival products. Any statement such as "X percent faster than product Y" cannot be supported by the Fact Pack.
The available quantitative performance information is contained in the raw specification fields. FP32 compute is listed at 3.333 TFLOPS, and FP16 compute is listed at 6.666 TFLOPS with a 2:1 ratio. This means the FP16 figure is exactly twice the FP32 figure, so workloads that can execute in FP16 have a numerically higher throughput ceiling than FP32-only paths. The pixel rate is 22.32 GPixel/s, and the texture rate is 52.08 GTexel/s. These fill-rate figures are tied to the 24 ROPs and the 56 TMUs in the design.
The memory bandwidth of 204.8 GB/s is the other major measured constraint. Combined with 32 GB of LPDDR5, it defines the memory-side behavior for any compute or rendering workload. The 40 W TDP is the power envelope within which the FP32 rate, FP16 rate, pixel rate, and texture rate are delivered.
In the absence of benchmark scores and rival entries, the only ranking evidence is percentileVsAllGpus = 50. That places the part at the median of the database's GPU distribution, but the zero average score and empty benchmark list mean this percentile is a relative placeholder rather than a result derived from measured performance. The raw throughput figures provide the substantive computational profile: 3.333 TFLOPS FP32, 6.666 TFLOPS FP16, 22.32 GPixel/s, and 52.08 GTexel/s. Without recorded scores, no further performance comparison can be quantified.
Detailed benchmark scores and charts for the NVIDIA Jetson AGX Orin 32 GB are below.
Benchmark Scores
No benchmark data available for this GPU.
Compare with Other GPUs
Select another GPU to compare specifications and benchmarks side-by-side.
Browse GPUs