NVIDIA B300 SXM6 AC vs NVIDIA GeForce RTX 5090 Mobile Comparison
NVIDIA B300 SXM6 AC
GeForce RTX 5090 Mobile
PERFORMANCE BENCHMARKS
Analysis: NVIDIA B300 SXM6 AC vs NVIDIA GeForce RTX 5090 Mobile
FAQ
Q: Which GPU has the higher Geekbench OpenCL score?
A: The NVIDIA B300 SXM6 AC scores 369831, while the NVIDIA GeForce RTX 5090 Mobile scores 201834. The B300 leads by 83.2% in this test.
Q: How does the B300 SXM6 AC compare to its nearest rivals?
A: The B300 sits at the 100th percentile among all GPUs. It is 7% ahead of the NVIDIA B200, 10.4% ahead of the NVIDIA H200 NVL, 16.3% ahead of the AMD Instinct MI300X, and 25% ahead of the NVIDIA L40S.
Q: Where does the RTX 5090 Mobile rank among all GPUs?
A: The RTX 5090 Mobile sits at the 84th percentile. Its average benchmark score is 45152, placing it near the AMD Radeon Pro 5500 XT (45384, 0.5% lower), the NVIDIA GeForce RTX 4070 Ti (44795, 0.8% higher), the Intel Arc A730M (45592, 1% lower), and the NVIDIA RTX 5880 Ada Generation (45972, 1.8% lower).
Q: What memory configurations do the two GPUs use?
A: The B300 SXM6 AC uses 288 GB of HBM3e on a 8192-bit bus with 8.19 TB/s bandwidth. The RTX 5090 Mobile uses 24 GB of GDDR7 on a 256-bit bus with 896.0 GB/s bandwidth.
Q: What are the power requirements of each card?
A: The B300 SXM6 AC has a TDP of 1100 W and a suggested PSU of 1500 W. The RTX 5090 Mobile has a TDP of 95 W and no suggested PSU listed.
Q: Which GPU supports DirectX and Vulkan?
A: The RTX 5090 Mobile supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The B300 SXM6 AC lists N/A for DirectX, OpenGL, and Vulkan, and has no display outputs.
Architecture Differences
The two GPUs come from different branches of NVIDIA's Blackwell family. The B300 SXM6 AC uses the GB110 chip built on the Blackwell Ultra architecture, belonging to the Server Blackwell (Bxx) generation. The RTX 5090 Mobile uses the GB203 chip built on Blackwell 2.0, belonging to the GeForce 50 Mobile generation.
Both chips are fabricated by TSMC on a 5 nm process, but the die sizes differ substantially. The GB110 measures 1628 mm² and packs 208,000 million transistors, yielding a transistor density of 127.8M per mm². The GB203 measures 378 mm² with 45,600 million transistors, for a density of 120.6M per mm². The B300's die is more than four times larger and carries roughly 4.5 times the transistor count.
Compute resources diverge sharply. The B300 has 18,944 shading units, 592 TMUs, 24 ROPs, and 592 tensor cores. The RTX 5090 Mobile has 10,496 shading units, 328 TMUs, 112 ROPs, 82 RT cores, and 328 tensor cores. The B300 does not list RT core count, while the mobile part includes dedicated ray tracing hardware.
Clock behavior also differs. The B300 runs at a 1665 MHz base and 2032 MHz boost. The RTX 5090 Mobile runs at a 990 MHz base and 1515 MHz boost. Memory clocks show 2000 MHz (8 Gbps effective) for the B300 versus 1750 MHz (28 Gbps effective) for the mobile GPU.
The B300 uses HBM3e memory with a massive 8192-bit bus, while the RTX 5090 Mobile uses GDDR7 on a 256-bit bus. The B300 is a SXM module with PCIe 6.0 x16 and no display outputs. The RTX 5090 Mobile is an IGP with PCIe 5.0 x16 and portable-device-dependent display outputs.
API support separates the two clearly. The B300 lists N/A for DirectX, OpenGL, and Vulkan, reflecting its server compute role. The RTX 5090 Mobile supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, making it suitable for gaming and graphics workloads.
Production status for both is Active. The B300 released on 2025-09-10, while the RTX 5090 Mobile released on 2025-03-26. The B300's predecessor is Server Hopper and its successor is Server Rubin. The RTX 5090 Mobile's predecessor is GeForce 40 Mobile, with no successor listed.
Head-to-Head Benchmarks
The database records only one common benchmark between these two GPUs: Geekbench OpenCL. The B300 SXM6 AC scores 369831, and the RTX 5090 Mobile scores 201834. The B300 wins by 83.2%, a dominant margin that reflects the massive difference in compute resources and memory bandwidth.
The B300's OpenCL score places it in the 100th percentile among all GPUs. Its nearest rivals trail by single-digit to double-digit percentages: the NVIDIA B200 at 345482 (7% lower), the NVIDIA H200 NVL at 334891 (10.4% lower), the AMD Instinct MI300X at 317994 (16.3% lower), and the NVIDIA L40S at 295763 (25% lower). The B300 leads every one of these data center parts by a clear margin.
The RTX 5090 Mobile's OpenCL score of 201834 is its strongest benchmark result in the database. Its other recorded tests show a broad range: 3DMark Steel Nomad DX12 at 5871, Geekbench Vulkan at 198405, Passmark G3D at 30034, and Passmark GPU Compute at 13401. DirectX-specific Passmark scores range from 138 (DX12) to 324 (DX9), with G2D at 1057.
The mobile GPU's average benchmark score is 45152, which is heavily weighted by its many Passmark results. Its nearest rivals cluster tightly around that average: the AMD Radeon Pro 5500 XT at 45384 (0.5% higher), the NVIDIA GeForce RTX 4070 Ti at 44795 (0.8% lower), the Intel Arc A730M at 45592 (1% higher), and the NVIDIA RTX 5880 Ada Generation at 45972 (1.8% higher). The RTX 5090 Mobile's percentile of 84 reflects this mid-pack positioning relative to the entire GPU database.
The single head-to-head result shows no ambiguity: the B300 wins the only shared test, and the database records 1 win for the B300 and 0 wins for the RTX 5090 Mobile.
Specification Differences
The specification sheets reveal stark contrasts in nearly every category.
The B300 uses the GB110 chip with Blackwell Ultra architecture, while the RTX 5090 Mobile uses the GB203 chip with Blackwell 2.0 architecture. The B300 belongs to the Server Blackwell (Bxx) generation; the RTX 5090 Mobile belongs to the GeForce 50 Mobile generation.
Transistor count: the B300 has 208,000 million transistors versus 45,600 million for the mobile GPU. Die size: 1628 mm² versus 378 mm². Transistor density: 127.8M per mm² versus 120.6M per mm².
Clocks: the B300 runs at 1665 MHz base and 2032 MHz boost, with memory at 2000 MHz (8 Gbps effective). The RTX 5090 Mobile runs at 990 MHz base and 1515 MHz boost, with memory at 1750 MHz (28 Gbps effective).
Memory: the B300 has 288 GB of HBM3e on a 8192-bit bus with 8.19 TB/s bandwidth. The RTX 5090 Mobile has 24 GB of GDDR7 on a 256-bit bus with 896.0 GB/s bandwidth.
Compute units: the B300 has 18,944 shading units, 592 TMUs, 24 ROPs, and 592 tensor cores. The RTX 5090 Mobile has 10,496 shading units, 328 TMUs, 112 ROPs, 82 RT cores, and 328 tensor cores. The B300 lists no RT core count.
Pixel and texture rates: the B300 achieves 48.77 GPixel/s and 1,202.9 GTexel/s. The RTX 5090 Mobile achieves 169.7 GPixel/s and 496.9 GTexel/s. The mobile GPU has a higher pixel rate due to its larger ROP count, while the B300 has a far higher texture rate.
FP32 and FP16: the B300 delivers 76.99 TFLOPS in both (1:1 ratio). The RTX 5090 Mobile delivers 31.80 TFLOPS in both (1:1 ratio).
Power: the B300 has a TDP of 1100 W with a suggested PSU of 1500 W. The RTX 5090 Mobile has a TDP of 95 W with no suggested PSU.
Form factor and interface: the B300 is a SXM Module with PCIe 6.0 x16 and no display outputs. The RTX 5090 Mobile is an IGP with PCIe 5.0 x16 and portable-device-dependent display outputs.
API support: the B300 lists N/A for DirectX, OpenGL, and Vulkan. The RTX 5090 Mobile lists DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.
Release dates: the B300 released on 2025-09-10; the RTX 5090 Mobile released on 2025-03-26.
Where Each One Wins
The B300 SXM6 AC wins decisively in compute throughput. Its OpenCL score of 369831 is 83.2% higher than the RTX 5090 Mobile's 201834. The B300's FP32 and FP16 figures of 76.99 TFLOPS more than double the mobile GPU's 31.80 TFLOPS. Its memory bandwidth of 8.19 TB/s is roughly nine times the mobile GPU's 896.0 GB/s, and its 288 GB capacity dwarfs the 24 GB on the RTX 5090 Mobile.
The B300 also wins on transistor count, die size, shading units, TMUs, tensor cores, texture rate, and bus width. It sits at the 100th percentile among all GPUs, leading every one of its nearest rivals by at least 7%. This is a server-class compute accelerator designed for workloads where raw number crunching and massive memory pools matter most.
The RTX 5090 Mobile wins in areas relevant to client graphics. Its pixel rate of 169.7 GPixel/s is more than triple the B300's 48.77 GPixel/s, thanks to 112 ROPs versus 24. It supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, while the B300 has no graphics API support. The mobile GPU has display outputs (portable-device-dependent), whereas the B300 has none.
The RTX 5090 Mobile also wins on power efficiency. Its 95 W TDP is dramatically lower than the B300's 1100 W, and it requires no external power connectors. The B300's suggested PSU of 1500 W indicates a system-level power demand that the mobile GPU avoids entirely.
In benchmark breadth, the RTX 5090 Mobile has ten recorded tests spanning 3DMark, Geekbench, and Passmark suites. The B300 has only one recorded test (Geekbench OpenCL). The mobile GPU's 84th percentile ranking shows it performs respectably across diverse workloads, even if its average score of 45152 places it near mid-range desktop parts like the RTX 4070 Ti.
The Verdict
The data points to two different missions. The B300 SXM6 AC is a data center compute accelerator. It offers 83.2% higher OpenCL performance than the RTX 5090 Mobile, sits at the 100th percentile, and leads all nearest rivals by at least 7%. Its 288 GB HBM3e memory, 8192-bit bus, and 76.99 TFLOPS FP32 make it suitable for large-scale AI training, scientific simulation, and similar workloads where power consumption of 1100 W is acceptable.
The RTX 5090 Mobile is a laptop GPU for graphics and gaming. It supports DirectX 12 Ultimate, Vulkan 1.4, and OpenGL 4.6, has display outputs, and delivers a higher pixel rate (169.7 GPixel/s) than the B300. Its 95 W TDP makes it practical for portable systems. Its benchmark profile shows balanced performance across 3DMark and Passmark tests, with an 84th percentile ranking among all GPUs.
For a builder choosing between these two, the decision rests on the use case. A server or workstation requiring maximum compute density and memory capacity should select the B300 SXM6 AC. A laptop or portable device needing graphics API support, display output, and reasonable power draw should select the RTX 5090 Mobile. The two share the Blackwell name and the 5 nm TSMC process, but little else in their specifications or benchmark results overlap.