AMD Radeon RX 9060 XT LP vs NVIDIA GB10 Comparison

The NVIDIA GB10 and AMD Radeon RX 9060 XT LP represent two very different approaches to GPU design, and the benchmark data shows a clear separation in performance and intended use. The GB10 is a server-oriented part with massive memory capacity, while the RX 9060 XT LP is a low-profile consumer graphics card. The recorded scores favor the GB10 decisively, but the architectural and specification differences explain why each product exists.

Head-to-Head Benchmarks

The Geekbench OpenCL test shows the NVIDIA GB10 scoring 120,137 points, while the AMD Radeon RX 9060 XT LP scores 88,183 points. This is a 36.2% advantage for the GB10. The OpenCL workload is a broad compute test, and the GB10’s higher shading unit count and server-class design drive this result. The RX 9060 XT LP is not weak, it sits in the 89th percentile of all GPUs, but the GB10 sits in the 95th percentile, a meaningful gap at the high end.

The Vulkan test produces an even larger separation. The GB10 scores 114,648, while the RX 9060 XT LP scores only 39,476. This is a 190.4% difference, effectively a landslide. Vulkan is a low-level graphics API, and the results indicate that the GB10 has substantially stronger low-level graphics execution capabilities, likely due to its architecture and driver optimization for compute-heavy workloads. The RX 9060 XT LP’s Vulkan score is unusually low relative to its OpenCL result, which suggests that its RDNA 4.0 architecture prioritizes DirectX 12 Ultimate and OpenGL paths, whereas the GB10 maintains high throughput across both APIs tested.

Looking at the average benchmark score, the GB10 achieves 117,393, while the RX 9060 XT LP averages 63,830. The nearest rivals for the GB10 include the NVIDIA RTX 4000 SFF Ada Generation at 117,088 (0.3% behind), the AMD Radeon PRO W7700 at 118,976 (1.3% ahead), and the NVIDIA Tesla V100 SXM2 16 GB at 114,395 (2.6% behind). The GB10 is tightly clustered with these professional and server cards, indicating that its performance class is workstation-grade. For the RX 9060 XT LP, the nearest rivals are much lower: the NVIDIA CMP 30HX at 63,842 (0% difference), the AMD Radeon RX 7600M at 63,775 (0.1% behind), and the AMD Radeon Pro Vega 56 at 63,693 (0.2% behind). This places the RX 9060 XT LP in the upper-midrange consumer segment, nowhere near the GB10’s tier.

The data shows that in both recorded benchmarks, the NVIDIA GB10 wins outright. The margin in Vulkan is particularly striking, nearly tripling the AMD card’s score. Even in OpenCL, where the RX 9060 XT LP is more competitive, the GB10 still leads by over a third.

The Verdict

From the recorded benchmark data, the NVIDIA GB10 is the superior performer in every measured test. Its OpenCL score is 36.2% higher, and its Vulkan score is 190.4% higher. The average benchmark score of 117,393 versus 63,830 reinforces this, a difference of roughly 84%. The GB10 also carries a 95th percentile ranking versus the RX 9060 XT LP’s 89th percentile, meaning it outperforms a larger share of the GPU population.

The GB10’s 128 GB of LPDDR5X memory is a decisive feature. No consumer card in this comparison comes close, and the 273.2 GB/s bandwidth, while not the highest, is paired with a 256-bit bus that supports server workloads requiring massive memory pools. The RX 9060 XT LP has 16 GB of GDDR6, which is ample for gaming and typical desktop tasks, but it is eight times smaller than the GB10’s capacity.

The RX 9060 XT LP does have a higher boost clock at 3050 MHz versus 2418 MHz, and a higher pixel rate at 195.2 GPixel/s versus 116.1 GPixel/s. However, these advantages do not translate into benchmark wins. The GB10’s raw shading power, 6144 shading units versus 2048, and its 384 tensor cores versus none on the AMD card, overwhelm the clock speed advantage. The verdict is clear: for raw compute and graphics throughput in the tested APIs, the GB10 is the definitive choice. The RX 9060 XT LP is a capable card, but it operates in a different performance class.

Where Each One Wins

The NVIDIA GB10 wins in compute-heavy scenarios. Its 29.71 TFLOPS of FP32 performance and 29.71 TFLOPS of FP16 performance, both at a 1:1 ratio, make it suitable for scientific simulation, AI inference, and data processing. The 384 tensor cores are specifically designed for matrix operations, a feature absent in the RX 9060 XT LP. The GB10 also wins in memory capacity, with 128 GB enabling datasets that would never fit in 16 GB. The PCIe 5.0 x16 interface on both cards is equal, but the GB10’s 5 nm process node and 382 mm² die size suggest a more complex, higher-yield design aimed at professional workloads.

The AMD Radeon RX 9060 XT LP wins in rasterization efficiency per clock. Its pixel rate of 195.2 GPixel/s is higher than the GB10’s 116.1 GPixel/s, meaning it can fill more screen pixels per second when clocked at its boost frequency. The texture rate of 390.4 GTexel/s is lower than the GB10’s 928.5 GTexel/s, so the AMD card does not win in texturing. The RX 9060 XT LP also has a higher memory bandwidth at 322.3 GB/s versus 273.2 GB/s, which helps in bandwidth-sensitive games. However, the Vulkan score suggests that the RX 9060 XT LP’s real-time graphics performance in that API is poor relative to its OpenCL showing, so its wins are limited to scenarios where DirectX 12 Ultimate is used, as it supports that API while the GB10 lists DirectX as N/A.

For users who need a low-profile card, the RX 9060 XT LP is a dual-slot design with a single 8-pin power connector, while the GB10 is an IGP (integrated graphics package) with no power connectors and a 150 mm length. The RX 9060 XT LP is the only one of the two that fits in a standard desktop expansion slot, making it the practical choice for gaming builds. The GB10 requires a server board that can accommodate an integrated package, which limits its applicability.

FAQ

Q: Which card has higher raw compute performance?

A: The NVIDIA GB10 delivers 29.71 TFLOPS of FP32 and FP16 performance, while the AMD Radeon RX 9060 XT LP delivers 24.99 TFLOPS in both. The GB10 leads by roughly 19% in raw floating-point throughput.

Q: How much memory does each card have?

A: The GB10 has 128 GB of LPDDR5X memory on a 256-bit bus with 273.2 GB/s bandwidth. The RX 9060 XT LP has 16 GB of GDDR6 on a 128-bit bus with 322.3 GB/s bandwidth.

Q: Why is the Vulkan score so different between the two?

A: In Geekbench Vulkan, the GB10 scores 114,648, while the RX 9060 XT LP scores 39,476, a 190.4% difference. This indicates the GB10 has far stronger Vulkan driver and hardware support, while the RX 9060 XT LP appears optimized for other APIs like DirectX 12 Ultimate.

Q: Are both cards currently in production?

A: Yes, both the NVIDIA GB10 and the AMD Radeon RX 9060 XT LP have an active production status according to the database.

Q: What are the power requirements?

A: Both cards have a 140 W TDP and a suggested PSU of 300 W. The GB10 uses no power connectors (it is an IGP), while the RX 9060 XT LP uses a single 8-pin connector.

Q: Which card has better API support?

A: The RX 9060 XT LP supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The GB10 lists DirectX, OpenGL, and Vulkan as N/A in the database, though its benchmark scores show it can execute Vulkan workloads.

Architecture Differences

The NVIDIA GB10 is built on the Blackwell 2.0 architecture, using the GB20B chip, while the AMD Radeon RX 9060 XT LP uses RDNA 4.0 with the Navi 44 chip. These are fundamentally different design philosophies. The GB10 is part of the Server Blackwell (Bxx) generation, targeting data center and professional compute. The RX 9060 XT LP belongs to the Navi IV (RX 9000) generation, aimed at consumer desktop graphics.

The process nodes differ: the GB10 uses a 5 nm process from TSMC, while the RX 9060 XT LP uses a 4 nm process from the same foundry. The smaller node on the AMD card allows for a higher transistor density of 149.2 million per square millimeter, but the GB10 has a larger die size of 382 mm² versus 199 mm². The RX 9060 XT LP has 29,700 million transistors, while the GB10’s transistor count is listed as unknown. The die size difference suggests the GB10 dedicates significant area to memory controllers and tensor cores.

The shading architecture is starkly different. The GB10 has 6144 shading units, 384 texture mapping units, and 48 ROPs. It also has 48 RT cores for ray tracing and 384 tensor cores for AI acceleration. The RX 9060 XT LP has 2048 shading units, 128 TMUs, and 64 ROPs, with 32 RT cores and no tensor cores. The GB10 has three times the shading units and three times the TMUs, but fewer ROPs (48 versus 64). The tensor cores are a major differentiator, as they enable the GB10 to handle neural network workloads that the RX 9060 XT LP cannot efficiently process.

The memory architecture also diverges. The GB10 uses LPDDR5X, a low-power memory type typically found in mobile and server applications, with a 256-bit bus. The RX 9060 XT LP uses GDDR6, a dedicated graphics memory, on a 128-bit bus. The AMD card has higher bandwidth (322.3 GB/s versus 273.2 GB/s) due to faster memory clocks, but the GB10 has four times the capacity.

The GB10’s pixel rate is 116.1 GPixel/s, lower than the RX 9060 XT LP’s 195.2 GPixel/s, but its texture rate of 928.5 GTexel/s is more than double the AMD card’s 390.4 GTexel/s. This indicates the GB10 is optimized for texture-heavy compute, while the AMD card is tuned for fill-rate-limited rasterization.

Specification Differences

The NVIDIA GB10 and AMD Radeon RX 9060 XT LP differ in nearly every measurable specification except TDP and bus interface. Both have a 140 W TDP and a suggested PSU of 300 W, and both use PCIe 5.0 x16. Beyond that, the differences are substantial.

The GB10 has a base clock of 1665 MHz and a boost clock of 2418 MHz, while the RX 9060 XT LP has a base clock of 1380 MHz, a game clock of 2450 MHz, and a boost clock of 3050 MHz. The AMD card clocks significantly higher, but this does not compensate for the GB10’s larger execution resources.

Memory size is the most obvious gap: 128 GB versus 16 GB. The memory type also differs, LPDDR5X versus GDDR6, as does the bus width, 256-bit versus 128-bit. Bandwidth favors the AMD card at 322.3 GB/s versus 273.2 GB/s, but capacity is the GB10’s overwhelming advantage.

The GB10 has 6144 shading units, 384 TMUs, 48 ROPs, 48 RT cores, and 384 tensor cores. The RX 9060 XT LP has 2048 shading units, 128 TMUs, 64 ROPs, and 32 RT cores, with no tensor cores. The GB10’s FP32 and FP16 performance is 29.71 TFLOPS, while the AMD card is 24.99 TFLOPS. Pixel rate favors the AMD card (195.2 GPixel/s versus 116.1 GPixel/s), as does memory bandwidth, but texture rate heavily favors the GB10 (928.5 GTexel/s versus 390.4 GTexel/s).

The physical form factors are entirely different. The GB10 is an IGP with no power connectors, measuring 150 mm by 51 mm by 150 mm. The RX 9060 XT LP is a dual-slot card with a single 8-pin power connector. The GB10 has one HDMI output, while the RX 9060 XT LP has one HDMI 2.1b and two DisplayPort 2.1a outputs. The GB10 does not list any API support (DirectX, OpenGL, Vulkan all N/A), while the RX 9060 XT LP supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4.

The release dates differ by two months, with the GB10 released on October 14, 2025, and the RX 9060 XT LP on December 16, 2025. The GB10 has a launch MSRP of 3,999 USD, which reflects its server positioning, while the RX 9060 XT LP does not have a recorded launch MSRP. The GB10’s predecessor is Server Hopper and its successor is Server Rubin, while the RX 9060 XT LP’s predecessor is Navi III and it has no listed successor. These specification differences confirm that the GB10 is a professional compute product, while the RX 9060 XT LP is a consumer graphics card with a focus on standard API compatibility and low-profile installation.