AMD Instinct MI308X vs Intel Arc Pro B60 Comparison

Head-to-Head Benchmarks

The database contains no shared benchmark scores for the AMD Instinct MI308X and Intel Arc Pro B60. The AMD part has no recorded benchmark entries, while the Intel card has eight tests in the database. A direct comparison of measured performance is therefore impossible from the recorded data. The two cards occupy entirely different segments, and their specifications reflect that divide.

The Intel Arc Pro B60 posts an average benchmark score of 3182 across its recorded tests. Its percentile rank versus all GPUs sits at 20, which places it below most of the database population. The nearest rivals in the database for the Intel card are low-end or older parts. The NVIDIA Quadro P1000 sits 0.6% ahead with an average score of 3163, while the GeForce GT 640 trails by 0.9% with 3210. The GeForce 920M is 3.2% behind with 3287, and the GeForce RTX 5080 SUPER is 3.5% ahead with 3075. These delta values are small, indicating that the Arc Pro B60 clusters closely with modest-performing GPUs in the database.

The AMD Instinct MI308X has no benchmark scores, no average score, no nearest rivals, and a percentile rank of 50. That percentile value is a placeholder rather than a measured outcome, since the card has zero recorded results. The data cannot establish where the MI308X stands relative to other accelerators. The absence of scores means no head-to-head wins can be assigned to either part. The wins counter for both cards is zero.

The Intel card's individual tests show a specific profile. Its highest single score is 14580 in PassMark G3D, followed by 7029 in PassMark GPU Compute. The DirectX tests are far lower: 179 in DirectX 9, 122 in DirectX 11, 76 in DirectX 12, and 61 in DirectX 10. The 2D test records 763. The 3DMark Steel Nomad DX12 test shows 2646. These numbers indicate a card that performs much better in general 3D workloads than in older DirectX API tests, which is consistent with a modern architecture designed for current graphics APIs.

The AMD card's theoretical compute figures are dramatically higher than the Intel card's measured scores, but those figures are not benchmarks. The MI308X lists FP32 at 81.72 TFLOPS, while the Arc Pro B60 lists FP32 at 12.29 TFLOPS. That is a 6.6x difference in raw floating-point throughput. Texture rate also diverges sharply: 2,553.6 GTexel/s for the AMD part versus 384.0 GTexel/s for the Intel part. Pixel rate tells a different story, as the MI308X reports 0 MPixel/s while the Intel card reports 192.0 GPixel/s. The AMD accelerator has no ROPs, which explains the zero pixel rate.

Memory bandwidth is another major gap. The MI308X delivers 5.32 TB/s from HBM3 across an 8192-bit bus, while the Arc Pro B60 delivers 456.0 GB/s from GDDR6 across a 192-bit bus. That is more than an 11x difference in bandwidth. The AMD card carries 192 GB of memory versus 24 GB on the Intel card, an 8x capacity difference.

FAQ

Q: Does the AMD Instinct MI308X have any benchmark scores in the database?

A: No. The MI308X has an empty benchmarks array, an average score of 0, and no nearest rivals. Its percentile rank of 50 is not derived from any recorded test.

Q: What is the Intel Arc Pro B60's best and worst recorded benchmark result?

A: The best score is 14580 in PassMark G3D, and the worst is 61 in PassMark DirectX 10.

Q: How does the Arc Pro B60 compare to its nearest rivals in the database?

A: It sits 0.6% ahead of the Quadro P1000, 0.9% behind the GeForce GT 640, 3.2% behind the GeForce 920M, and 3.5% behind the GeForce RTX 5080 SUPER.

Q: Which card has higher FP32 throughput?

A: The AMD Instinct MI308X lists 81.72 TFLOPS, which is 6.6 times the Arc Pro B60's 12.29 TFLOPS.

Q: Which card has more memory bandwidth?

A: The MI308X has 5.32 TB/s from HBM3, versus 456.0 GB/s from GDDR6 on the Arc Pro B60. The AMD card's bandwidth is roughly 11.7 times higher.

Q: Which card supports display outputs?

A: The Intel Arc Pro B60 has 4x mini-DisplayPort 2.1 outputs. The AMD Instinct MI308X has no display outputs.

The Verdict

The data supports a clear segmentation between these two products, but it does not support a performance comparison. The MI308X has no recorded benchmarks, so any assertion about its speed relative to the Arc Pro B60 would be speculation. The only verifiable differences are in specifications and in the Intel card's measured scores.

For compute-heavy workloads, the MI308X's specifications point to a purpose-built accelerator. It uses HBM3 memory with 5.32 TB/s bandwidth, 192 GB capacity, and an 8192-bit bus. Its FP32 throughput of 81.72 TFLOPS and texture rate of 2,553.6 GTexel/s are far beyond the Intel card's figures. The MI308X has no display outputs and no graphics API support, which confirms it is not intended for rendering to a screen. The card requires a 1150 W suggested PSU and uses an OAM module form factor with no power connectors on the board itself.

The Arc Pro B60 is a conventional graphics card. It has 24 GB of GDDR6 memory, 456.0 GB/s bandwidth, and a 192-bit bus. Its FP32 figure is 12.29 TFLOPS, and its pixel rate is 192.0 GPixel/s. It supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, and it offers 4x mini-DisplayPort 2.1 outputs. The card uses a dual-slot design with a single 8-pin power connector and a 550 W suggested PSU. Its dimensions are 167 mm in length, 69 mm in height, and 40 mm in width.

The benchmark data for the Arc Pro B60 indicates modest performance relative to the rest of the database. Its 20th percentile rank and the small delta values against low-end rivals show that it does not compete with high-end consumer or workstation GPUs in the recorded tests. The MI308X cannot be placed on that scale because it has no scores.

The choice between these two cards depends entirely on the intended use case. A system that needs graphics output, standard API support, and a compact dual-slot card would use the Arc Pro B60. A system that needs massive memory capacity, extreme bandwidth, and high compute throughput without display output would use the MI308X. The data does not indicate that either card is superior in general; they serve different functions.

Specification Differences

The two cards differ across nearly every specification field. The MI308X uses HBM3 memory with 192 GB capacity, an 8192-bit bus, and 5.32 TB/s bandwidth. The Arc Pro B60 uses GDDR6 with 24 GB capacity, a 192-bit bus, and 456.0 GB/s bandwidth.

The MI308X has 19456 shading units, 1216 TMUs, and 0 ROPs. The Arc Pro B60 has 2560 shading units, 160 TMUs, and 80 ROPs. The Intel card also has 20 RT cores, while the AMD card lists no RT cores. The MI308X reports 0 MPixel/s pixel rate, while the Arc Pro B60 reports 192.0 GPixel/s. Texture rate is 2,553.6 GTexel/s for the AMD card and 384.0 GTexel/s for the Intel card.

Clock speeds differ. The MI308X has a 1000 MHz base and 2100 MHz boost, with memory at 1300 MHz (5.2 Gbps effective). The Arc Pro B60 has a 2000 MHz base and 2400 MHz boost, with memory at 2375 MHz (19 Gbps effective).

Power and physical specifications diverge. The MI308X has a 750 W TDP, a suggested PSU of 1150 W, an OAM Module slot width, and no power connectors. The Arc Pro B60 has a 200 W TDP, a suggested PSU of 550 W, a dual-slot width, and a single 8-pin connector. The bus interface is PCIe 5.0 x16 on the AMD card and PCIe 5.0 x8 on the Intel card.

The AMD card has no display outputs. The Intel card has 4x mini-DisplayPort 2.1. The MI308X lists no API support, while the Arc Pro B60 supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4. The MI308X has no recorded dimensions, while the Arc Pro B60 measures 167 mm by 69 mm by 40 mm.

Architecture Differences

The MI308X uses the Aqua Vanjaram chip with CDNA 3.0 architecture, part of the Instinct (MIx) generation. The Arc Pro B60 uses the BMG-G21 chip with Xe2-HPG architecture, part of the Battlemage (Pro Series) generation.

Both cards are built on a 5 nm process at TSMC, but the transistor counts differ significantly. The MI308X has 153,000 million transistors across a 1017 mm² die, giving a density of 150.4M per mm². The Arc Pro B60 has 19,600 million transistors across a 272 mm² die, giving a density of 72.1M per mm². The AMD chip is roughly 3.7 times larger in die area and carries nearly 8 times more transistors.

The memory architectures reflect different design goals. The MI308X uses HBM3, which provides extremely high bandwidth in a stacked configuration, while the Arc Pro B60 uses GDDR6 on a conventional board. The AMD card's 8192-bit bus is 42.7 times wider than the Intel card's 192-bit bus.

Compute feature sets differ. The MI308X has no RT cores, while the Arc Pro B60 has 20. The MI308X lists FP16 at 81.72 TFLOPS (1:1 ratio with FP32), whereas the Arc Pro B60 lists FP16 at 24.58 TFLOPS (2:1 ratio with FP32).

The release dates are far apart. The MI308X launched on 2023-12-05, while the Arc Pro B60 launched on 2025-09-04. The Intel card has an active production status, and the AMD card has no recorded production status. The MI308X lists a predecessor of Radeon Instinct, while the Arc Pro B60 has no predecessor. The MI308X has a launch MSRP of none, while the Arc Pro B60 has a launch MSRP of 499 USD.