Intel Arc B370 vs NVIDIA H200 NVL Comparison

Intel Arc B370 is an integrated graphics processor built on Intel's Panther Lake chip, using the Xe3-LPG architecture and a 3 nm process. The NVIDIA H200 NVL is a server accelerator built on the GH100 chip, using the Hopper architecture and a 5 nm TSMC process. The database records a single benchmark for each product, and neither product has shared head-to-head benchmark results. The Arc B370 scores 1184 in 3DMark Steel Nomad DX12, placing it in the 5th percentile of all GPUs. The H200 NVL scores 334891 in Geekbench OpenCL, placing it in the 100th percentile of all GPUs.

Where Each One Wins

The data shows a complete separation in workload categories. The Intel Arc B370 wins in graphics-focused tasks, specifically DirectX 12 gaming workloads, as evidenced by its 3DMark Steel Nomad score. This is a low-power integrated solution designed for portable devices, with a 25 W TDP and no discrete power connectors. Its architecture supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, making it a functional option for light gaming and general graphics acceleration on a mobile platform.

The NVIDIA H200 NVL wins decisively in compute-intensive workloads. Its Geekbench OpenCL score of 334891 reflects a massive compute capability, driven by 16896 shading units, 528 tensor cores, and 60.32 TFLOPS of FP32 performance. The H200 NVL is a dual-slot server card with 141 GB of HBM3e memory and 4.89 TB/s of bandwidth, designed for large-scale parallel processing, AI inference, and scientific computing. It has no display outputs, confirming its role as a pure compute accelerator rather than a graphics card.

Benchmark results indicate that the H200 NVL outperforms the Arc B370 by a factor of roughly 283x in their respective recorded benchmarks, though they target completely different segments. The Arc B370 sits in the 5th percentile globally, while the H200 NVL sits in the 100th percentile. Each product wins in its intended domain: the Arc B370 for integrated graphics on a 25 W platform, the H200 NVL for high-throughput server compute.

FAQ

Q: How do the two products compare in raw benchmark scores?

A: The Intel Arc B370 records a 3DMark Steel Nomad DX12 score of 1184, placing it in the 5th percentile of all GPUs. The NVIDIA H200 NVL records a Geekbench OpenCL score of 334891, placing it in the 100th percentile of all GPUs.

Q: What are the nearest rivals for each product based on the database?

A: The Arc B370's nearest rivals include the ATI Mobility Radeon HD 5570 (score 1186, delta -0.2%), ATI Radeon HD 5770 (score 1190, delta -0.5%), AMD Radeon HD 7650M (score 1192, delta -0.7%), and AMD FirePro M2000 (score 1168, delta +1.4%). The H200 NVL's nearest rivals include the NVIDIA B200 (score 345482, delta -3.1%), AMD Instinct MI300X (score 317994, delta +5.3%), NVIDIA B300 SXM6 AC (score 369831, delta -9.4%), and NVIDIA L40S (score 295763, delta +13.2%).

Q: What is the memory configuration of each product?

A: The Arc B370 uses system shared memory, with the memory type, bus width, and bandwidth all listed as system dependent. The H200 NVL has 141 GB of HBM3e memory on a 6144-bit bus, delivering 4.89 TB/s of bandwidth.

Q: What are the power requirements for each product?

A: The Arc B370 has a 25 W TDP, uses no power connectors, and is an integrated graphics processor (IGP). The H200 NVL has a 600 W TDP, requires an 8-pin EPS power connector, and has a suggested power supply of 1000 W.

Q: Which product supports DirectX 12?

A: The Intel Arc B370 supports DirectX 12 Ultimate (12_2), along with OpenGL 4.6 and Vulkan 1.4. The NVIDIA H200 NVL lists DirectX, OpenGL, and Vulkan support as N/A, indicating it does not target consumer graphics APIs.

Q: What is the production status of each product?

A: Both products are listed as Active in the database. The Arc B370 was released on 2026-01-26, while the H200 NVL was released on 2024-11-17.

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark results between the Intel Arc B370 and the NVIDIA H200 NVL. The wins tally is 0 for each product in shared tests. However, the individual benchmark scores provide the only comparable metric.

In 3DMark Steel Nomad DX12, the Arc B370 scores 1184. This places it just 0.2% below the ATI Mobility Radeon HD 5570 (1186) and 0.5% below the ATI Radeon HD 5770 (1190). It is 0.7% behind the AMD Radeon HD 7650M (1192) and 1.4% ahead of the AMD FirePro M2000 (1168). These figures show the Arc B370 competing within a narrow performance band around 1180-1190 points.

In Geekbench OpenCL, the H200 NVL scores 334891. This is 3.1% behind the NVIDIA B200 (345482) and 9.4% behind the NVIDIA B300 SXM6 AC (369831). It is 5.3% ahead of the AMD Instinct MI300X (317994) and 13.2% ahead of the NVIDIA L40S (295763). The H200 NVL sits in the upper tier of server accelerators, with only the B200 and B300 above it in the recorded data.

The performance gap between the two products is enormous. The H200 NVL's OpenCL score is roughly 283 times higher than the Arc B370's 3DMark score, though different benchmarks measure different capabilities. The Arc B370 targets mobile graphics with a 25 W TDP, while the H200 NVL targets server compute with a 600 W TDP. The data shows no overlap in their performance envelopes.

Specification Differences

The Intel Arc B370 and NVIDIA H200 NVL differ in nearly every measurable specification. The Arc B370 has a base clock of 300 MHz and a boost clock of 2400 MHz, while the H200 NVL has a base clock of 1365 MHz and a boost clock of 1785 MHz. The Arc B370's memory is system shared, while the H200 NVL uses 141 GB of HBM3e with a 6144-bit bus.

The shading unit count differs substantially: the Arc B370 has 1280 shading units, 40 texture mapping units, and 20 raster output units, while the H200 NVL has 16896 shading units, 528 texture mapping units, and 24 raster output units. The Arc B370 has 10 ray tracing cores and no tensor cores, while the H200 NVL has no listed ray tracing cores but has 528 tensor cores.

Pixel and texture rates also diverge. The Arc B370 delivers 48.00 GPixel/s and 96.00 GTexel/s, while the H200 NVL delivers 42.84 GPixel/s and 942.5 GTexel/s. The FP32 compute is 6.144 TFLOPS for the Arc B370 versus 60.32 TFLOPS for the H200 NVL. FP16 performance is 12.29 TFLOPS (2:1) for the Arc B370 and 120.6 TFLOPS (2:1) for the H200 NVL.

Physical specifications differ completely. The Arc B370 is an integrated processor with a slot width of IGP, no power connectors, and a 25 W TDP. The H200 NVL is a dual-slot card measuring 267 mm in length and 111 mm in height, requiring an 8-pin EPS connector, a 600 W TDP, and a suggested 1000 W power supply. The Arc B370 uses an IGP bus interface, while the H200 NVL uses PCIe 5.0 x16. The Arc B370's display outputs are portable device dependent, while the H200 NVL has no display outputs.

Architecture Differences

The two products use entirely different architectures from different manufacturers. The Intel Arc B370 uses the Xe3-LPG architecture on a 3 nm process, built on the Panther Lake chip, and belongs to the Arc Graphics-M (Panther Lake) generation. The NVIDIA H200 NVL uses the Hopper architecture on a 5 nm process, built on the GH100 chip, and belongs to the Server Hopper (Hxx) generation. The foundry for the Arc B370 is Intel, while the H200 NVL uses TSMC.

The H200 NVL has a known transistor count of 80,000 million on a die size of 814 mm², giving a transistor density of 98.3 million transistors per square millimeter. The Arc B370's transistor count and die size are listed as unknown in the database. The H200 NVL's predecessor is Server Ada and its successor is Server Blackwell, while the Arc B370 has no listed predecessor or successor.

The memory architecture differs fundamentally. The Arc B370 relies on system shared memory, meaning its bandwidth is system dependent. The H200 NVL uses dedicated HBM3e memory with 141 GB capacity, 6144-bit bus width, and 4.89 TB/s bandwidth, running at 1593 MHz with 6.4 Gbps effective data rate.

API support also separates the two. The Arc B370 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The H200 NVL lists all three APIs as N/A, confirming its role as a compute-focused accelerator without consumer graphics API support. The Arc B370 has 10 ray tracing cores for hardware-accelerated ray tracing, while the H200 NVL has no listed ray tracing cores but includes 528 tensor cores for AI and matrix operations.

The Verdict

The recorded data indicates that the Intel Arc B370 and NVIDIA H200 NVL serve completely different purposes, and the choice between them depends entirely on the workload. The Arc B370 is an integrated graphics solution for portable devices, with a 25 W TDP, system shared memory, and support for DirectX 12 Ultimate. Its 3DMark Steel Nomad score of 1184 places it near the ATI Mobility Radeon HD 5570 and ATI Radeon HD 5770, confirming its position as a low-power mobile graphics option.

The NVIDIA H200 NVL is a server-class compute accelerator with a 600 W TDP, 141 GB of HBM3e memory, and 4.89 TB/s of bandwidth. Its Geekbench OpenCL score of 334891 places it in the 100th percentile of all GPUs, and it sits within 3.1% of the NVIDIA B200 and 9.4% of the NVIDIA B300 SXM6 AC. This product targets large-scale parallel compute, AI workloads, and scientific applications, with no display outputs and no consumer graphics API support.

For a portable device requiring integrated graphics, the Arc B370 is the only viable option from this pair. For a server requiring maximum compute throughput, the H200 NVL delivers performance that the Arc B370 cannot approach. The data shows a performance ratio of roughly 283x in favor of the H200 NVL in their respective benchmarks, but this comparison is not apples-to-apples. The Arc B370 wins in the mobile integrated graphics segment, and the H200 NVL wins in the server accelerator segment. There is no overlap in their intended use cases.