Intel Arc Pro A60 vs NVIDIA B300 SXM6 AC Comparison

Head-to-Head Benchmarks

The database records a single shared benchmark result for these two accelerators: Geekbench OpenCL. The NVIDIA B300 SXM6 AC posts a score of 369,831, while the Intel Arc Pro A60 records 63,485. The delta is -82.8% in favor of the NVIDIA part. In absolute terms, the B300 SXM6 AC delivers roughly 5.8 times the OpenCL performance of the Arc Pro A60. This is not a marginal gap; it is a fundamental separation in compute throughput class.

The B300 SXM6 AC sits at the 100th percentile among all GPUs in the database, meaning no recorded device scores higher. Its average benchmark score is identical to its single OpenCL result at 369,831. The Arc Pro A60, by contrast, holds the 88th percentile, with an average benchmark score of 60,326 across its two recorded tests (OpenCL at 63,485 and Vulkan at 57,166). The Vulkan score has no direct counterpart on the B300, which reports no Vulkan result, so the only head-to-head comparison is OpenCL.

Looking at nearest rivals, the B300 SXM6 AC leads the NVIDIA B200 by 7%, the H200 NVL by 10.4%, the AMD Instinct MI300X by 16.3%, and the L40S by 25%. These deltas show that the B300 does not merely edge out its competition; it establishes a clear tier of its own. The Intel Arc Pro A60, meanwhile, trades within a narrow band around the 60,000-point mark: it is within 0.3% of the AMD Radeon Pro Vega 48, trails the AMD Radeon Pro W6600M by 2.5%, and leads the AMD Radeon PRO V710 by 2.8%. Its nearest rival by average score is the NVIDIA GeForce RTX 4090, which records 60,347, a 0% delta against the Arc Pro A60's 60,326.

The single head-to-head result is decisive. The B300 SXM6 AC wins the only recorded comparison by a wide margin, and the data shows no benchmark category where the Arc Pro A60 posts a higher score. The wins tally reflects this: 0 wins for the Intel part, 1 win for the NVIDIA part.

Where Each One Wins

The B300 SXM6 AC wins in raw compute throughput across every measurable dimension in the database. Its FP32 output is 76.99 TFLOPS, compared to 8.397 TFLOPS for the Arc Pro A60, a 9.2-fold difference. FP16 performance is 76.99 TFLOPS (1:1 ratio) on the NVIDIA part versus 16.79 TFLOPS (2:1 ratio) on the Intel part. The B300 also leads in texture rate at 1,202.9 GTexel/s against 262.4 GTexel/s, and in memory bandwidth at 8.19 TB/s versus 384.0 GB/s. These are not marginal improvements; they are order-of-magnitude differences that place the B300 in a different performance class.

The Arc Pro A60 wins in areas unrelated to raw compute speed but relevant to specific deployment contexts. It has display outputs: four DisplayPort 2.0 connectors, while the B300 SXM6 AC has no outputs at all. The Intel part is a single-slot design with a 130 W TDP and a suggested PSU of 300 W, while the B300 is an SXM module with a 1100 W TDP and a suggested PSU of 1500 W. The Arc Pro A60 also supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4; the B300 reports N/A for all three APIs. For graphics workloads, rendering, or any task requiring a display output, the Arc Pro A60 is the only viable option in this comparison. For compute workloads, the B300 is the only viable option if maximum throughput is the goal.

The B300's pixel rate is 48.77 GPixel/s, which is lower than the Arc Pro A60's 131.2 GPixel/s. This is an unusual inversion: the B300 has far more shading units (18,944 versus 2,048) and TMUs (592 versus 128) but only 24 ROPs versus 64 on the Intel part. The data indicates the B300 is not optimized for rasterization-style pixel output, consistent with its server-oriented, no-display design. The Arc Pro A60's higher pixel rate and lower power envelope make it suited for workstation graphics tasks, while the B300's immense FP32, FP16, and memory bandwidth make it suited for large-scale compute, AI training, and inference workloads.

Architecture Differences

The two accelerators come from different foundries and process nodes. The Intel Arc Pro A60 uses a DG2-256 chip on a 6 nm TSMC process, with 11,500 million transistors on a 269 mm² die, yielding a transistor density of 42.8 million per mm². The NVIDIA B300 SXM6 AC uses a GB110 chip on a 5 nm TSMC process, with 208,000 million transistors on a 1,628 mm² die, yielding a density of 127.8 million per mm². The B300 packs roughly 18 times more transistors into a die that is about 6 times larger physically, and its density is 3 times higher per square millimeter.

Architecturally, the Intel part is Xe-HPG, generation Alchemist (Pro Series), while the NVIDIA part is Blackwell Ultra, generation Server Blackwell (Bxx). The Intel chip has 2,048 shading units, 128 TMUs, 64 ROPs, and 16 ray tracing cores. It has no tensor cores listed. The NVIDIA chip has 18,944 shading units, 592 TMUs, 24 ROPs, and 592 tensor cores. It lists no RT cores. The B300's FP16 throughput matches its FP32 at a 1:1 ratio, indicating native FP16 capability, whereas the Arc Pro A60 achieves FP16 at a 2:1 ratio, meaning it halves throughput when using FP16 versus FP32.

Memory configurations differ sharply. The Arc Pro A60 uses 12 GB of GDDR6 on a 192-bit bus, delivering 384.0 GB/s bandwidth. The B300 uses 288 GB of HBM3e on an 8192-bit bus, delivering 8.19 TB/s. The B300's memory bus is 42.7 times wider, and its bandwidth is 21.3 times higher. Clock speeds are closer: the Arc Pro A60 has a base of 900 MHz and a boost of 2050 MHz, while the B300 has a base of 1665 MHz and a boost of 2032 MHz. Memory clocks are both listed at 2000 MHz, with the Intel part at 16 Gbps effective and the NVIDIA part at 8 Gbps effective.

The bus interfaces differ as well: the Arc Pro A60 uses PCIe 4.0 x16, while the B300 uses PCIe 6.0 x16. The B300 has no display outputs and no graphics API support, while the Intel part supports modern graphics APIs and multiple display connections. Production status for both is Active. The B300 was released on 2025-09-10, with a predecessor listed as Server Hopper and a successor as Server Rubin. The Arc Pro A60 was released on 2023-06-05, with no predecessor or successor listed.

The Verdict

The data describes two products with almost no functional overlap. The NVIDIA B300 SXM6 AC is a server accelerator with no display outputs, no graphics API support, and an 1100 W power profile. Its 76.99 TFLOPS FP32 and 8.19 TB/s memory bandwidth place it at the 100th percentile, and it leads its nearest rivals by 7% to 25%. Any workload that requires maximum FP32, FP16, or memory throughput will see a decisive advantage from the B300. Its 288 GB of HBM3e memory and 592 tensor cores make it suited for large model training, inference, and scientific compute.

The Intel Arc Pro A60 is a workstation graphics card with four DisplayPort 2.0 outputs, DirectX 12 Ultimate support, OpenGL 4.6, and Vulkan 1.4. It draws 130 W and fits in a single slot. Its 12 GB GDDR6 and 384.0 GB/s bandwidth are modest by comparison, but its pixel rate of 131.2 GPixel/s exceeds the B300's 48.77 GPixel/s. The Arc Pro A60 sits at the 88th percentile, with an average score of 60,326, and trades within 2.8% of several AMD workstation parts.

For users who need a displayable graphics card for rendering, CAD, or any task involving a monitor, the Arc Pro A60 is the only option with the required outputs and API support. For users who need raw compute density, the B300 SXM6 AC is the only option that delivers the recorded performance levels. The 82.8% OpenCL deficit of the Arc Pro A60 against the B300 cannot be overcome by any configuration change. The verdict from the data is straightforward: choose the Arc Pro A60 for graphics workstations and the B300 for server compute. Neither device is a substitute for the other.

FAQ

Q: What is the OpenCL score difference between the Intel Arc Pro A60 and the NVIDIA B300 SXM6 AC?

A: The Intel Arc Pro A60 scores 63,485 in Geekbench OpenCL, while the NVIDIA B300 SXM6 AC scores 369,831. The delta is -82.8%, meaning the B300 leads by a factor of about 5.8.

Q: Does the Intel Arc Pro A60 have any advantage in pixel throughput?

A: Yes. The Arc Pro A60 has a pixel rate of 131.2 GPixel/s, while the B300 SXM6 AC has a pixel rate of 48.77 GPixel/s. This aligns with the Intel part having 64 ROPs versus 24 on the NVIDIA part.

Q: What memory configurations do the two accelerators use?

A: The Intel Arc Pro A60 uses 12 GB of GDDR6 on a 192-bit bus with 384.0 GB/s bandwidth. The NVIDIA B300 SXM6 AC uses 288 GB of HBM3e on an 8192-bit bus with 8.19 TB/s bandwidth.

Q: Which device supports graphics APIs and display outputs?

A: The Intel Arc Pro A60 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, and has four DisplayPort 2.0 outputs. The NVIDIA B300 SXM6 AC reports N/A for all graphics APIs and has no display outputs.

Q: How does the B300 SXM6 AC compare to its nearest rivals in the database?

A: The B300 SXM6 AC leads the NVIDIA B200 by 7%, the NVIDIA H200 NVL by 10.4%, the AMD Instinct MI300X by 16.3%, and the NVIDIA L40S by 25%. Its average benchmark score is 369,831, at the 100th percentile.

Q: What are the FP32 and FP16 performance figures for each device?

A: The Intel Arc Pro A60 delivers 8.397 TFLOPS FP32 and 16.79 TFLOPS FP16 (2:1 ratio). The NVIDIA B300 SXM6 AC delivers 76.99 TFLOPS FP32 and 76.99 TFLOPS FP16 (1:1 ratio).