Intel Arc Graphics 128EU Mobile vs NVIDIA B200 SXM6 Comparison

Head-to-Head Benchmarks

The benchmark database contains no direct head-to-head measurements between the Intel Arc Graphics 128EU Mobile and the NVIDIA B200 SXM6. Both products have an average benchmark score of 0 and hold the 50th percentile among all GPUs in the database. This absence of recorded comparative data is expected, as these two devices occupy entirely different segments of the market. The Intel part is an integrated graphics solution for mobile processors, while the NVIDIA part is a server-class accelerator module. Without direct measurements, the comparison must rely on their architectural specifications and the roles those specifications imply.

The raw compute difference is substantial. The NVIDIA B200 SXM6 delivers 69.34 TFLOPS of FP32 performance, which is roughly 15 times the 4.608 TFLOPS of the Intel Arc Graphics 128EU Mobile. In FP16 workloads, the gap narrows slightly in relative terms but remains enormous in absolute numbers: the B200 delivers 69.34 TFLOPS at 1:1 ratio, while the Intel part reaches 9.216 TFLOPS using a 2:1 ratio. The B200 also carries 592 tensor cores, a feature class entirely absent from the Intel Arc Graphics 128EU Mobile specification sheet.

Texture and pixel throughput tell a similar story. The B200's texture rate of 1,083.4 GTexel/s is over seven times the Intel part's 144.0 GTexel/s. Interestingly, the Intel Arc Graphics 128EU Mobile has a higher pixel rate at 72.00 GPixel/s compared to the B200's 43.92 GPixel/s, despite the B200's overwhelming advantage in other metrics. This is a consequence of the B200's unusually low ROP count of 24 relative to its other resources, while the Intel part carries 32 ROPs.

Memory bandwidth is where the B200 separates itself most decisively. The B200 uses 180 GB of HBM3e across an 8192-bit bus, yielding 8.19 TB/s of bandwidth. The Intel Arc Graphics 128EU Mobile relies on system shared memory, with bandwidth described as system dependent. No fixed bandwidth number exists for the Intel part, so a direct numerical comparison cannot be made. The architectural difference, however, is categorical: dedicated HBM3e versus shared system memory.

FAQ

Q: Which GPU has more shading units?

A: The NVIDIA B200 SXM6 has 18,944 shading units. The Intel Arc Graphics 128EU Mobile has 1,024. The B200 also has 592 texture mapping units and 592 tensor cores, while the Intel part has 64 TMUs and no tensor cores.

Q: What are the power requirements for each?

A: The Intel Arc Graphics 128EU Mobile has a TDP of 28 W. The NVIDIA B200 SXM6 has a TDP of 1000 W and a suggested PSU rating of 1400 W.

Q: Do both GPUs support DirectX?

A: No. The Intel Arc Graphics 128EU Mobile supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The NVIDIA B200 SXM6 lists N/A for DirectX, OpenGL, and Vulkan, indicating no consumer graphics API support.

Q: What memory types do these GPUs use?

A: The Intel Arc Graphics 128EU Mobile uses system shared memory with a system dependent bus width and bandwidth. The NVIDIA B200 SXM6 uses 180 GB of HBM3e memory on an 8192-bit bus with 8.19 TB/s bandwidth.

Q: When were these products released?

A: The Intel Arc Graphics 128EU Mobile was released on December 13, 2023. The NVIDIA B200 SXM6 was released on October 31, 2024.

Q: What process nodes are used?

A: The Intel Arc Graphics 128EU Mobile uses a 10 nm process at Intel foundry. The NVIDIA B200 SXM6 uses a 5 nm process at TSMC, with 208,000 million transistors on a 1628 mm² die.

Architecture Differences

The Intel Arc Graphics 128EU Mobile is built on the Meteor Lake chip using the Xe-LPG architecture, placed in the Arc Graphics-M generation. It uses a 10 nm process at Intel's own foundry. The NVIDIA B200 SXM6 is built on the GB100 chip using the Blackwell architecture, belonging to the Server Blackwell generation, fabricated on a 5 nm process at TSMC. The B200 packs 208,000 million transistors onto a 1628 mm² die, achieving a transistor density of 127.8 million transistors per mm². The Intel part's transistor count and die size are not recorded in the database.

The Intel Arc Graphics 128EU Mobile has a base clock of 300 MHz and a boost clock of 2250 MHz. The NVIDIA B200 SXM6 has a much lower base clock of 120 MHz but a boost clock of 1830 MHz. Clock speeds alone do not explain the performance gap; the B200's advantage comes from having nearly 18.5 times the shading units and a far wider memory interface.

The B200 includes 592 tensor cores, a feature the Intel part does not list. The B200 also implements 592 TMUs compared to the Intel part's 64. ROP counts are closer: 32 on the Intel part and 24 on the B200, which explains the Intel part's higher pixel rate of 72.00 GPixel/s versus 43.92 GPixel/s.

The B200 uses HBM3e memory with 180 GB capacity, 8192-bit bus width, and 8.19 TB/s bandwidth. The Intel Arc Graphics 128EU Mobile uses system shared memory with system dependent bandwidth, meaning its memory performance scales with the host system's memory configuration rather than a dedicated pool.

The B200's API support is listed as N/A for DirectX, OpenGL, and Vulkan. This is consistent with its role as a compute-oriented server accelerator with no display outputs. The Intel part supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, and its display outputs are described as portable device dependent, reflecting its mobile integration role.

The B200 uses a PCIe 6.0 x16 bus interface and comes as an SXM Module. The Intel Arc Graphics 128EU Mobile uses a Ring Bus interface and is packaged as an IGP, meaning it is integrated into the processor package. The B200 has no display outputs. The Intel part's output capability depends on the portable device it is built into.

The B200 has a TDP of 1000 W and a suggested PSU of 1400 W. The Intel Arc Graphics 128EU Mobile has a TDP of 28 W. The B200's predecessor is Server Hopper, and its successor is Server Rubin. The Intel part's predecessor is HD Graphics-M, and no successor is listed.

The Verdict

The data shows two products with no meaningful overlap in purpose. The Intel Arc Graphics 128EU Mobile is an integrated GPU for laptops and portable devices. It uses system shared memory, has a 28 W TDP, supports consumer graphics APIs, and produces display output. Its strengths are low power draw, compact integration as an IGP, and standard graphics feature support.

The NVIDIA B200 SXM6 is a server accelerator module. It has 180 GB of HBM3e, 8.19 TB/s of memory bandwidth, 69.34 TFLOPS of FP32 compute, and 592 tensor cores. It has no display outputs and no consumer API support. Its 1000 W TDP and 1400 W suggested PSU place it firmly in datacenter territory.

The benchmark data contains no head-to-head results, and the percentile rankings are identical at 50. The only recorded specification overlap is the production status: both are listed as Active. The B200 launch MSRP is 34,999 USD. The Intel part has no recorded launch MSRP.

For a user choosing between these two, the decision is not about performance per watt or benchmark scores. It is about the role of the hardware. The Intel Arc Graphics 128EU Mobile exists to provide graphics capability inside a mobile processor package. The NVIDIA B200 SXM6 exists to accelerate compute workloads in servers. Nothing in the database suggests either could substitute for the other.

Specification Differences

The following fields differ between the Intel Arc Graphics 128EU Mobile and the NVIDIA B200 SXM6:

• Chip: Meteor Lake versus GB100
• Architecture: Xe-LPG versus Blackwell
• Generation: Arc Graphics-M (Meteor Lake) versus Server Blackwell (Bxx)
• Process node: 10 nm versus 5 nm
• Foundry: Intel versus TSMC
• Transistors: not recorded versus 208,000 million
• Die size: not recorded versus 1628 mm²
• Transistor density: not recorded versus 127.8M per mm²
• Base clock: 300 MHz versus 120 MHz
• Boost clock: 2250 MHz versus 1830 MHz
• Memory type: System Shared versus HBM3e
• Memory size: System Shared versus 180 GB
• Memory bus width: System Shared versus 8192 bit
• Memory bandwidth: System Dependent versus 8.19 TB/s
• Shading units: 1024 versus 18,944
• TMUs: 64 versus 592
• ROPs: 32 versus 24
• Tensor cores: none listed versus 592
• Pixel rate: 72.00 GPixel/s versus 43.92 GPixel/s
• Texture rate: 144.0 GTexel/s versus 1,083.4 GTexel/s
• FP32 performance: 4.608 TFLOPS versus 69.34 TFLOPS
• FP16 performance: 9.216 TFLOPS (2:1) versus 69.34 TFLOPS (1:1)
• TDP: 28 W versus 1000 W
• Slot width: IGP versus SXM Module
• Suggested PSU: none listed versus 1400 W
• Bus interface: Ring Bus versus PCIe 6.0 x16
• Display outputs: Portable Device Dependent versus No outputs
• DirectX support: 12 (12_1) versus N/A
• OpenGL support: 4.6 versus N/A
• Vulkan support: 1.4 versus N/A
• Release date: December 13, 2023 versus October 31, 2024
• Predecessor: HD Graphics-M versus Server Hopper
• Successor: none listed versus Server Rubin
• Launch MSRP: none recorded versus 34,999 USD

Where Each One Wins

The Intel Arc Graphics 128EU Mobile wins in power efficiency by a wide margin. Its 28 W TDP is a fraction of the B200's 1000 W. It also has a higher pixel rate at 72.00 GPixel/s versus 43.92 GPixel/s, a direct consequence of its 32 ROPs against the B200's 24. The Intel part supports consumer graphics APIs including DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, making it suitable for standard graphics workloads. Its display outputs, described as portable device dependent, mean it can drive screens in mobile devices. Its Ring Bus interface suits its integrated role.

The NVIDIA B200 SXM6 wins decisively in compute throughput. Its 69.34 TFLOPS FP32 performance is roughly 15 times the Intel part. Its FP16 performance of 69.34 TFLOPS at 1:1 ratio surpasses the Intel part's 9.216 TFLOPS at 2:1 ratio. The B200's 8.19 TB/s memory bandwidth from 180 GB of HBM3e is in a different class from system shared memory. Its 592 tensor cores provide dedicated hardware for tensor operations that the Intel part lacks entirely. The B200 also has a higher texture rate at 1,083.4 GTexel/s versus 144.0 GTexel/s.

The use case split follows the data. The Intel Arc Graphics 128EU Mobile suits portable devices that need integrated graphics with low power draw, consumer API support, and display output. The NVIDIA B200 SXM6 suits server environments that need massive FP32 and FP16 throughput, large HBM3e memory capacity, tensor core acceleration, and no display functionality. The B200's PCIe 6.0 x16 interface and SXM Module form factor align with server motherboard integration. The Intel part's IGP form factor and Ring Bus interface align with mobile processor packages.

The B200's 208,000 million transistors and 1628 mm² die indicate a design optimized for maximum compute density. The Intel part's 10 nm process, while larger than the B200's 5 nm node, suits a low-power integrated part. The B200's transistor density of 127.8M per mm² reflects the TSMC 5 nm process. Neither product records any benchmark scores in the database, so the verdict rests entirely on the specification record.