Intel Arc Pro B390 vs NVIDIA B300 SXM6 AC Comparison

Intel Arc Pro B390 and NVIDIA B300 SXM6 AC occupy opposite ends of the hardware spectrum, and the recorded data reflects that divide. The B300 SXM6 AC is a server-class accelerator with a dominant benchmark score, while the Arc Pro B390 is an integrated graphics processor designed for portable devices. The database shows a clear performance hierarchy, but the comparison also highlights fundamental architectural and specification differences that define their respective roles.

Head-to-Head Benchmarks

The benchmark data is decisively one-sided. The NVIDIA B300 SXM6 AC delivers a Geekbench OpenCL score of 369,831, placing it in the 100th percentile of all GPUs in the database. This is an exceptional result, confirming its position as a top-tier compute accelerator. In contrast, the Intel Arc Pro B390 has no recorded benchmark scores in the database, resulting in an average score of 0 and a 50th percentile ranking.

The B300 SXM6 AC’s performance advantage over its nearest rivals is substantial. It leads the NVIDIA B200 by 7 percent, the NVIDIA H200 NVL by 10.4 percent, the AMD Instinct MI300X by 16.3 percent, and the NVIDIA L40S by 25 percent. These delta percentages, derived from the database’s average scores, show that the B300 SXM6 AC is not merely competitive but sits at the top of the server accelerator hierarchy. For example, the B200 scores 345,482, and the H200 NVL scores 334,891, both well behind the B300 SXM6 AC’s 369,831.

The Intel Arc Pro B390’s lack of benchmark data means no direct head-to-head score comparison is possible. However, the sheer scale of the B300 SXM6 AC’s compute resources, such as its 76.99 TFLOPS FP32 throughput, dwarfs the Arc Pro B390’s 7.680 TFLOPS. The database records a 10-fold difference in raw FP32 performance, which translates directly into the benchmark results. In practical terms, the B300 SXM6 AC is built for massive parallel workloads, while the Arc Pro B390 targets lightweight, integrated use cases.

FAQ

Q: How does the NVIDIA B300 SXM6 AC compare to its closest rivals?

A: The B300 SXM6 AC records an OpenCL score of 369,831, which is 7 percent higher than the NVIDIA B200, 10.4 percent higher than the NVIDIA H200 NVL, 16.3 percent higher than the AMD Instinct MI300X, and 25 percent higher than the NVIDIA L40S. It sits in the 100th percentile of all GPUs in the database.

Q: What is the memory configuration of the B300 SXM6 AC?

A: The B300 SXM6 AC uses 288 GB of HBM3e memory with an 8192-bit bus width, providing 8.19 TB/s of bandwidth. The memory clock runs at 2000 MHz with 8 Gbps effective speed.

Q: Does the Intel Arc Pro B390 have any benchmark results?

A: No, the database lists no benchmark scores for the Arc Pro B390. Its average benchmark score is 0, and it holds a 50th percentile ranking among all GPUs, which is a default position rather than a measured performance level.

Q: What are the API capabilities of the Intel Arc Pro B390?

A: The Arc Pro B390 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The NVIDIA B300 SXM6 AC lists N/A for DirectX, OpenGL, and Vulkan, as it targets compute workloads rather than graphics rendering.

Q: What is the process node and foundry for each chip?

A: The Intel Arc Pro B390 uses a 3 nm process node fabricated by Intel. The NVIDIA B300 SXM6 AC uses a 5 nm process node fabricated by TSMC.

Q: How do the power requirements differ?

A: The Intel Arc Pro B390 has a TDP of 80 W and requires no power connectors, as it is an integrated graphics processor. The NVIDIA B300 SXM6 AC has a TDP of 1100 W and lists a suggested PSU of 1500 W.

Architecture Differences

The two processors are built on entirely different architectural foundations. The Intel Arc Pro B390 uses the Xe3-LPG architecture based on the Panther Lake chip, part of the Arc Graphics-WM generation. It is manufactured on a 3 nm process at Intel’s foundry. The NVIDIA B300 SXM6 AC uses the Blackwell Ultra architecture based on the GB110 chip, part of the Server Blackwell (Bxx) generation, and is manufactured on a 5 nm process at TSMC.

The Intel chip integrates 1,536 shading units, 48 texture mapping units, and 24 raster operations pipelines. It also includes 12 ray tracing cores, making it suitable for graphics workloads that require hardware-accelerated ray tracing. The NVIDIA chip, by contrast, packs 18,944 shading units, 592 texture mapping units, and 592 tensor cores. It does not list dedicated ray tracing cores, reflecting its focus on AI and compute tasks rather than real-time graphics.

The transistor counts reveal the scale difference. The B300 SXM6 AC contains 208,000 million transistors on a 1628 mm² die, with a transistor density of 127.8M per mm². The Arc Pro B390’s transistor count and die size are listed as unknown in the database. This asymmetry underscores the B300 SXM6 AC’s role as a massive data center part, while the Arc Pro B390 is a compact integrated solution.

Memory architecture diverges sharply as well. The Arc Pro B390 uses system shared memory, with bandwidth described as system dependent. This means it relies on the host system’s RAM and has no dedicated VRAM. The B300 SXM6 AC uses 288 GB of HBM3e with a dedicated 8192-bit bus, delivering 8.19 TB/s of bandwidth. This dedicated, high-bandwidth memory is essential for the B300 SXM6 AC’s compute-intensive server workloads.

Specification Differences

The database records several fields where the two parts differ completely. The process node differs: 3 nm for Intel versus 5 nm for NVIDIA. The foundry differs: Intel for the Arc Pro B390, TSMC for the B300 SXM6 AC. The base clock differs significantly: 300 MHz for the Intel part versus 1665 MHz for the NVIDIA part. The boost clock also differs: 2500 MHz for Intel versus 2032 MHz for NVIDIA.

Memory specifications are entirely different. The Arc Pro B390 lists system shared memory for size, type, bus width, and bandwidth, with bandwidth described as system dependent. The B300 SXM6 AC specifies 288 GB of HBM3e, an 8192-bit bus, and 8.19 TB/s of bandwidth. The shading units differ: 1,536 for Intel versus 18,944 for NVIDIA. Texture mapping units differ: 48 versus 592. Raster operations pipelines are the same at 24, but the ray tracing cores (12 for Intel, none listed for NVIDIA) and tensor cores (none for Intel, 592 for NVIDIA) differ.

Pixel rate and texture rate differ as well. The Arc Pro B390 records a pixel rate of 60.00 GPixel/s and a texture rate of 120.0 GTexel/s. The B300 SXM6 AC records a pixel rate of 48.77 GPixel/s and a texture rate of 1,202.9 GTexel/s. FP32 performance is 7.680 TFLOPS for Intel versus 76.99 TFLOPS for NVIDIA. FP16 performance is 15.36 TFLOPS (2:1) for Intel versus 76.99 TFLOPS (1:1) for NVIDIA.

Power and physical specifications differ starkly. The Arc Pro B390 has a TDP of 80 W, a slot width of IGP, no power connectors, and a bus interface of IGP. The B300 SXM6 AC has a TDP of 1100 W, a slot width of SXM Module, and a PCIe 6.0 x16 bus interface. The Intel part has display outputs described as portable device dependent, while the NVIDIA part lists no outputs. API support also differs: Intel supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, while NVIDIA lists N/A for all three.

Where Each One Wins

The NVIDIA B300 SXM6 AC wins in every compute-oriented category. Its FP32 throughput of 76.99 TFLOPS is 10 times the Intel part’s 7.680 TFLOPS. Its FP16 performance of 76.99 TFLOPS (1:1) exceeds Intel’s 15.36 TFLOPS (2:1) by a factor of five. The B300 SXM6 AC’s texture rate of 1,202.9 GTexel/s is an order of magnitude higher than the Arc Pro B390’s 120.0 GTexel/s. The 288 GB HBM3e memory with 8.19 TB/s bandwidth is designed for data center scale workloads, whereas the Intel part’s system shared memory is inherently limited by the host system.

The Intel Arc Pro B390 wins in power efficiency and integration. Its 80 W TDP allows it to function as an IGP with no power connectors, making it suitable for portable devices where power draw and physical space are constrained. The B300 SXM6 AC requires a 1500 W suggested PSU and uses an SXM Module slot, which is impractical outside a server chassis. The Arc Pro B390 also supports modern graphics APIs, including DirectX 12 Ultimate and Vulkan 1.4, making it a functional graphics solution for embedded or mobile displays. The B300 SXM6 AC offers no display outputs and no graphics API support, confirming it is a compute accelerator, not a rendering device.

The pixel rate is one area where the Intel part leads: 60.00 GPixel/s versus 48.77 GPixel/s for NVIDIA. This indicates that the Arc Pro B390 can drive pixel-heavy workloads at a higher rate despite its lower overall compute, which is relevant for display output tasks in portable devices.

The release timeline also differs. The B300 SXM6 AC was released on September 10, 2025, and is listed as active with a successor, Server Rubin. The Arc Pro B390 was released on January 26, 2026, and is also active, with a predecessor of HD Graphics-WM. The B300 SXM6 AC’s predecessor is Server Hopper, and its successor is Server Rubin, placing it in a clear generational line of NVIDIA server products.

In summary, the B300 SXM6 AC dominates raw compute, memory bandwidth, and tensor performance, making it the clear choice for AI training, scientific simulation, and large-scale data processing. The Arc Pro B390 excels as a low-power integrated graphics solution, offering modern API support and adequate pixel throughput for portable device displays, but it cannot compete in any compute benchmark where the B300 SXM6 AC operates. The database’s percentile rankings reflect this: the B300 SXM6 AC sits at 100, while the Arc Pro B390 sits at 50 with no measured scores.