Intel Arc Pro B370 vs NVIDIA B300 SXM6 AC Comparison

FAQ

Q: What are the two products compared here?

A: The comparison is between the Intel Arc Pro B370, an integrated graphics processor (IGP) based on the Panther Lake chip with Xe3-LPG architecture, and the NVIDIA B300 SXM6 AC, a server accelerator module based on the GB110 chip with Blackwell Ultra architecture.

Q: Which product has a higher benchmark percentile ranking?

A: The NVIDIA B300 SXM6 AC sits at the 100th percentile against all GPUs in the database, while the Intel Arc Pro B370 sits at the 50th percentile. The NVIDIA part also has a recorded average benchmark score of 369,831, while the Intel part has no recorded benchmark entries.

Q: What is the memory configuration of each product?

A: The Intel Arc Pro B370 uses system shared memory with a system dependent bandwidth and a system shared bus width. The NVIDIA B300 SXM6 AC has 288 GB of HBM3e memory on an 8192-bit bus with 8.19 TB/s of bandwidth.

Q: What are the power requirements for each product?

A: The Intel Arc Pro B370 has a 25 W TDP and uses no power connectors, functioning as an integrated processor. The NVIDIA B300 SXM6 AC has a 1100 W TDP and requires a suggested PSU of 1500 W, mounted as an SXM module.

Q: Which product supports DirectX 12 Ultimate?

A: The Intel Arc Pro B370 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The NVIDIA B300 SXM6 AC has no API support listed, with DirectX, OpenGL, and Vulkan all marked as N/A, and it has no display outputs.

Q: What are the production statuses and release dates?

A: Both products are listed as Active in production status. The Intel Arc Pro B370 was released on January 26, 2026, with a predecessor of HD Graphics-WM. The NVIDIA B300 SXM6 AC was released on September 10, 2025, with a predecessor of Server Hopper and a successor of Server Rubin.

Architecture Differences

The Intel Arc Pro B370 and NVIDIA B300 SXM6 AC represent fundamentally different design philosophies. The Intel part is built on a 3 nm process at Intel's foundry, using the Panther Lake chip with Xe3-LPG architecture. It belongs to the Arc Graphics-WM (Panther Lake) generation. The NVIDIA B300 SXM6 AC is fabricated on a 5 nm process at TSMC, using the GB110 chip with Blackwell Ultra architecture, and belongs to the Server Blackwell (Bxx) generation.

The transistor counts and die sizes could not be more different. The Intel part has an unknown transistor count and die size, while the NVIDIA part packs 208,000 million transistors onto a 1628 mm² die, achieving a transistor density of 127.8 million transistors per square millimeter. This massive scale difference reflects the target markets: the Intel part is an integrated processor with no slot width beyond IGP, while the NVIDIA part is a dedicated SXM module.

The NVIDIA B300 SXM6 AC incorporates 592 tensor cores, a feature entirely absent from the Intel Arc Pro B370's specification sheet. The Intel part instead lists 10 ray tracing cores, while the NVIDIA part does not list RT core counts. The memory architecture also diverges sharply: the Intel part uses system shared memory with system dependent bandwidth, while the NVIDIA part uses dedicated HBM3e memory with a colossal 8192-bit bus width.

The NVIDIA part has a base clock of 1665 MHz and a boost clock of 2032 MHz, with memory running at 2000 MHz (8 Gbps effective). The Intel part has a base clock of just 300 MHz and a boost clock of 2400 MHz. The NVIDIA part's interface is PCIe 6.0 x16, while the Intel part uses an IGP bus interface with no power connectors required.

Where Each One Wins

The Intel Arc Pro B370 wins in scenarios requiring integrated graphics with minimal power draw. Its 25 W TDP makes it suitable for portable devices, as its display outputs are listed as "Portable Device Dependent." It supports modern graphics APIs including DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, making it a viable option for systems needing API compatibility without a discrete GPU.

The NVIDIA B300 SXM6 AC wins decisively in raw compute performance. Its 76.99 TFLOPS of FP32 performance dwarfs the Intel part's 6.144 TFLOPS. The NVIDIA part also delivers 76.99 TFLOPS of FP16 performance at a 1:1 ratio, while the Intel part offers 12.29 TFLOPS of FP16 at a 2:1 ratio. The NVIDIA part has no display outputs, indicating it is designed purely for server compute workloads rather than graphics rendering.

For memory bandwidth, the NVIDIA B300 SXM6 AC is in a completely different class with 8.19 TB/s of bandwidth versus the Intel part's system dependent figure. The NVIDIA part's 288 GB of HBM3e memory provides massive capacity for large datasets, while the Intel part relies on shared system memory.

The NVIDIA part also wins on texture and pixel throughput. Its texture rate is 1,202.9 GTexel/s versus the Intel part's 96.00 GTexel/s, and its pixel rate is 48.77 GPixel/s versus 48.00 GPixel/s. In shading units, the NVIDIA part has 18,944 versus the Intel part's 1,280, and its 592 TMUs compare to the Intel part's 40.

Specification Differences

The two products differ across nearly every specification category. The process node differs: 3 nm for Intel versus 5 nm for NVIDIA. The foundry differs: Intel for the Arc Pro B370, TSMC for the B300 SXM6 AC. Transistor counts are unknown for Intel but listed at 208,000 million for NVIDIA, with a die size of 1628 mm².

Clock speeds differ substantially. The Intel part runs at 300 MHz base and 2400 MHz boost, while the NVIDIA part runs at 1665 MHz base and 2032 MHz boost. Memory clocks also differ: the Intel part uses system shared memory, while the NVIDIA part uses 2000 MHz with 8 Gbps effective.

Compute unit counts diverge completely. The Intel Arc Pro B370 has 1,280 shading units, 40 TMUs, 20 ROPs, and 10 RT cores, with no tensor cores listed. The NVIDIA B300 SXM6 AC has 18,944 shading units, 592 TMUs, 24 ROPs, and 592 tensor cores, with no RT cores listed.

The power envelope is dramatically different: 25 W for Intel versus 1100 W for NVIDIA. The suggested PSU is only listed for the NVIDIA part at 1500 W. The slot width is IGP for Intel and SXM Module for NVIDIA. Bus interfaces are IGP versus PCIe 6.0 x16. Display outputs are "Portable Device Dependent" for Intel and "No outputs" for NVIDIA.

API support is exclusive to the Intel part, with DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4 all listed as N/A for the NVIDIA part. The release dates differ by roughly four months, with NVIDIA releasing September 10, 2025 and Intel releasing January 26, 2026.

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark entries between these two products, and the win counts show zero for each side. However, the NVIDIA B300 SXM6 AC has a recorded Geekbench OpenCL score of 369,831, placing it at the 100th percentile of all GPUs. The Intel Arc Pro B370 has no recorded benchmark scores, with an average benchmark score of zero and a 50th percentile ranking.

The NVIDIA B300 SXM6 AC's nearest rivals in the database provide context for its performance. It sits 7% ahead of the NVIDIA B200, which has an average score of 345,482. It is 10.4% ahead of the NVIDIA H200 NVL, which scores 334,891. Against the AMD Instinct MI300X, it leads by 16.3%, with that rival scoring 317,994. The largest gap is against the NVIDIA L40S, where the B300 SXM6 AC is 25% ahead of the 295,763 score.

The FP32 compute figures reinforce this dominance. The NVIDIA B300 SXM6 AC delivers 76.99 TFLOPS, which is 12.5 times the Intel Arc Pro B370's 6.144 TFLOPS. In FP16, the NVIDIA part delivers 76.99 TFLOPS at 1:1 ratio, while the Intel part delivers 12.29 TFLOPS at 2:1 ratio, making the NVIDIA advantage roughly 6.3 times.

Texture rate comparisons show the NVIDIA part at 1,202.9 GTexel/s versus the Intel part's 96.00 GTexel/s, a 12.5 times advantage. Pixel rates are closer, with the NVIDIA part at 48.77 GPixel/s and the Intel part at 48.00 GPixel/s, representing only a 1.6% difference. The shading unit count ratio is 14.8 to 1 in favor of NVIDIA, and TMU count is 14.8 to 1 as well.

The Verdict

The data indicates two products serving entirely separate markets with no overlap in use cases. The Intel Arc Pro B370 is an integrated graphics solution for portable devices, drawing only 25 W, using system shared memory, and supporting modern graphics APIs. It has no benchmark scores recorded in the database and sits at the 50th percentile.

The NVIDIA B300 SXM6 AC is a server accelerator with a 1100 W TDP, 288 GB of HBM3e memory, and 8.19 TB/s of bandwidth. Its 76.99 TFLOPS FP32 performance places it at the 100th percentile, and its nearest rivals trail by 7% to 25%. It has no display outputs and no graphics API support, confirming its role as a compute-only product.

For systems requiring integrated graphics with low power consumption and API compatibility, the Intel Arc Pro B370 is the appropriate choice. For server workloads demanding maximum FP32 and FP16 throughput, massive memory capacity, and the highest benchmark percentile ranking, the NVIDIA B300 SXM6 AC is the clear selection. The 50th versus 100th percentile ranking difference, combined with the absence of any recorded benchmarks for the Intel part, shows that the database contains no evidence of competitive overlap between these two products.