Intel Arc A380M vs Intel Arc Pro B65 Comparison

Intel Arc A380M and Intel Arc Pro B65 represent two distinct points in Intel’s GPU lineup, separated by architecture generation, process technology, and performance class. The A380M is a mobile-focused MXM module built on the Alchemist architecture, while the Pro B65 is a dual-slot desktop workstation card based on the newer Battlemage architecture. The database shows no direct head-to-head benchmark entries for these two parts, and neither product carries recorded average benchmark scores or percentile ranks beyond a shared 50th percentile placement among all GPUs. Because of this, the analysis relies on the architectural and specification data recorded for each unit, which reveals substantial differences in compute throughput, memory capacity, bandwidth, and power requirements.

Head-to-Head Benchmarks

The recorded data contains no head-to-head benchmark results between the Intel Arc A380M and the Intel Arc Pro B65. The database lists zero wins for either product in direct comparison, and no benchmark entries appear for either GPU. This absence of measured performance data means a direct frame-rate or compute-score comparison cannot be constructed from the available information. What can be compared are the theoretical throughput figures recorded in the specification fields, which provide a clear picture of the relative capability of each design.

The most significant gap appears in FP32 compute. The Intel Arc Pro B65 delivers 12.29 TFLOPS of single-precision performance, while the Intel Arc A380M delivers 4.096 TFLOPS. The Pro B65 therefore provides three times the FP32 throughput of the A380M, a margin that places the two cards in different performance classes entirely. In FP16 compute, the Pro B65 reaches 24.58 TFLOPS, compared to 8.192 TFLOPS for the A380M, again a threefold advantage. Both architectures use a 2:1 ratio for FP16 relative to FP32, so the relative scaling remains consistent across both precision formats.

Texture and pixel throughput follow the same pattern. The Pro B65 records a texture rate of 384.0 GTexel/s, exactly three times the 128.0 GTexel/s of the A380M. Pixel rate for the Pro B65 is 192.0 GPixel/s, also exactly three times the 64.00 GPixel/s of the A380M. This consistency across FP32, FP16, texture, and pixel metrics indicates the Pro B65 is not merely a higher-clocked version of the A380M, but a fundamentally larger implementation with more processing resources at every stage of the pipeline.

Memory bandwidth shows an even larger disparity. The Pro B65 records 608.0 GB/s of bandwidth through a 256-bit memory bus, while the A380M records 186.0 GB/s through a 96-bit bus. The Pro B65 offers more than three times the memory bandwidth of the A380M, which directly affects workloads that are sensitive to memory throughput, such as large dataset processing, high-resolution texture streaming, and compute tasks with high data reuse. The memory capacity difference is also substantial: 32 GB on the Pro B65 versus 6 GB on the A380M, a factor of more than five.

Clock speeds favor the Pro B65 as well, though the differences are less extreme than the throughput gaps. The Pro B65 runs at a base clock of 2400 MHz and a boost clock of 2400 MHz, meaning the card operates at a fixed frequency with no boost delta. The A380M runs at a base clock of 1550 MHz and a boost clock of 2000 MHz, so its boost clock is 450 MHz higher than its base, but still 400 MHz below the Pro B65’s fixed clock. Memory clock also favors the Pro B65 at 2375 MHz with 19 Gbps effective data rate, versus 1937 MHz with 15.5 Gbps effective on the A380M.

Where Each One Wins

Without benchmark results, the recorded specifications define the areas where each product holds an advantage. The Intel Arc Pro B65 wins in every raw performance metric listed in the database. Its FP32 compute is three times higher, its texture fill rate is three times higher, its pixel fill rate is three times higher, its memory bandwidth is more than three times higher, and its memory capacity is more than five times higher. The Pro B65 also has more than double the shading units, 2560 versus 1024, and more than double the ray tracing cores, 20 versus 8. It has 160 texture mapping units versus 64, and 80 render output units versus 32. Every processing resource that contributes to rendering or compute throughput is larger on the Pro B65.

The Intel Arc A380M holds advantages in power consumption and physical footprint. The A380M is rated at 35 W TDP, while the Pro B65 is rated at 200 W TDP. The A380M uses an MXM module form factor, specifically MXM-A (3.1), and its display output is described as portable device dependent, meaning it is designed for integration into mobile or compact systems. The Pro B65 is a dual-slot card using a PCIe 5.0 x16 interface, consumes significantly more power, requires a single 8-pin power connector, and carries a suggested power supply rating of 550 W. For systems constrained by space or power delivery, the A380M is the only viable option between the two.

The architecture generation also separates the two in terms of feature baseline. The A380M uses the Xe-HPG architecture from the Alchemist generation, while the Pro B65 uses Xe2-HPG from the Battlemage generation. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so the API feature set recorded in the database is identical. The Pro B65 provides four DisplayPort 2.1 outputs, which supports high-resolution multi-monitor workstation setups, while the A380M relies on the host device for display connectivity.

Architecture Differences

The two GPUs are built on different architectures and different process nodes. The A380M uses the DG2-128 chip with the Xe-HPG architecture, belonging to the Alchemist generation, specifically the Arc 3 Mobile segment. The Pro B65 uses the BMG-G21 chip with the Xe2-HPG architecture, belonging to the Battlemage generation, specifically the Pro Series. This architectural jump from Alchemist to Battlemage represents a significant generational shift in Intel’s GPU design, with the Xe2-HPG architecture delivering higher resource counts and higher clock frequencies per watt in the recorded data.

Process technology differs as well. The A380M is fabricated on a 6 nm process at TSMC, while the Pro B65 is fabricated on a 5 nm process at TSMC. The Pro B65 uses a more advanced process node, which contributes to its ability to reach 2400 MHz clocks despite carrying a much larger chip. The A380M chip contains 7,200 million transistors on a die size of 157 mm², giving a transistor density of 45.9 million transistors per square millimeter. The Pro B65 chip contains 19,600 million transistors on a die size of 272 mm², giving a transistor density of 72.1 million transistors per square millimeter. The Pro B65 packs nearly three times the transistor count into less than twice the die area, reflecting both the larger design and the denser process node.

The memory subsystems differ in both width and capacity. The A380M uses 6 GB of GDDR6 on a 96-bit bus, with 186.0 GB/s of bandwidth. The Pro B65 uses 32 GB of GDDR6 on a 256-bit bus, with 608.0 GB/s of bandwidth. The bus width difference of 96 bits versus 256 bits is the primary driver of the bandwidth gap, while the capacity difference of 6 GB versus 32 GB reflects the workstation positioning of the Pro B65. Both use GDDR6 memory, so the memory type is identical, but the effective data rate differs: 15.5 Gbps on the A380M versus 19 Gbps on the Pro B65.

The processing resource counts scale in proportion to the chip size difference. The A380M has 1024 shading units, 64 TMUs, 32 ROPs, and 8 ray tracing cores. The Pro B65 has 2560 shading units, 160 TMUs, 80 ROPs, and 20 ray tracing cores. The Pro B65 has 2.5 times the shading units, 2.5 times the TMUs, 2.5 times the ROPs, and 2.5 times the ray tracing cores. This 2.5x scaling across all processing units indicates a consistent design multiplier rather than a partial upgrade in only one stage of the pipeline. The clock advantage of the Pro B65, 2400 MHz versus 2000 MHz boost on the A380M, combines with this 2.5x resource scaling to produce the 3x throughput advantage observed in the FP32, texture, and pixel rate figures.

Power delivery and board design also reflect the different target platforms. The A380M is an MXM module with no recorded power connector, relying on the host system for power. The Pro B65 is a dual-slot card with a single 8-pin power connector and a suggested power supply of 550 W. The TDP figures of 35 W for the A380M and 200 W for the Pro B65 place them at opposite ends of the power spectrum, with the Pro B65 consuming nearly six times the power of the A380M. The bus interface also differs: MXM-A (3.1) for the A380M versus PCIe 5.0 x16 for the Pro B65, which aligns with the mobile versus desktop workstation positioning.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The Intel Arc Pro B65 delivers 12.29 TFLOPS of FP32 compute, while the Intel Arc A380M delivers 4.096 TFLOPS. The Pro B65 provides three times the single-precision throughput of the A380M.

Q: How much memory does each GPU have?

A: The Intel Arc A380M has 6 GB of GDDR6 on a 96-bit bus, while the Intel Arc Pro B65 has 32 GB of GDDR6 on a 256-bit bus. The Pro B65 also records higher bandwidth at 608.0 GB/s versus 186.0 GB/s.

Q: What are the power requirements for each card?

A: The Intel Arc A380M is rated at 35 W TDP and uses an MXM module form factor. The Intel Arc Pro B65 is rated at 200 W TDP, uses a dual-slot design, requires a single 8-pin power connector, and has a suggested power supply of 550 W.

Q: Do both GPUs support the same APIs?

A: Yes, both the Intel Arc A380M and the Intel Arc Pro B65 support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The API feature sets recorded in the database are identical.

Q: What architecture does each GPU use?

A: The Intel Arc A380M uses the Xe-HPG architecture with the DG2-128 chip from the Alchemist generation. The Intel Arc Pro B65 uses the Xe2-HPG architecture with the BMG-G21 chip from the Battlemage generation.

Q: Which GPU has more ray tracing cores?

A: The Intel Arc Pro B65 has 20 ray tracing cores, while the Intel Arc A380M has 8 ray tracing cores. The Pro B65 has 2.5 times the ray tracing core count of the A380M.

Specification Differences

The table below lists only the fields where the two recorded products differ.

| Field | Intel Arc A380M | Intel Arc Pro B65 |

|-------|-----------------|-------------------|

| Chip | DG2-128 | BMG-G21 |

| Architecture | Xe-HPG | Xe2-HPG |

| Generation | Alchemist (Arc 3 Mobile) | Battlemage (Pro Series) |

| Process Node | 6 nm | 5 nm |

| Transistors | 7,200 million | 19,600 million |

| Die Size | 157 mm² | 272 mm² |

| Transistor Density | 45.9M / mm² | 72.1M / mm² |

| Base Clock | 1550 MHz | 2400 MHz |

| Boost Clock | 2000 MHz | 2400 MHz |

| Memory Clock | 1937 MHz, 15.5 Gbps effective | 2375 MHz, 19 Gbps effective |

| Memory Size | 6 GB | 32 GB |

| Memory Bus Width | 96 bit | 256 bit |

| Memory Bandwidth | 186.0 GB/s | 608.0 GB/s |

| Shading Units | 1024 | 2560 |

| TMUs | 64 | 160 |

| ROPs | 32 | 80 |

| Ray Tracing Cores | 8 | 20 |

| Pixel Rate | 64.00 GPixel/s | 192.0 GPixel/s |

| Texture Rate | 128.0 GTexel/s | 384.0 GTexel/s |

| FP32 Performance | 4.096 TFLOPS | 12.29 TFLOPS |

| FP16 Performance | 8.192 TFLOPS (2:1) | 24.58 TFLOPS (2:1) |

| TDP | 35 W | 200 W |

| Slot Width | MXM Module | Dual-slot |

| Power Connectors | None recorded | 1x 8-pin |

| Suggested PSU | None recorded | 550 W |

| Bus Interface | MXM-A (3.1) | PCIe 5.0 x16 |

| Display Outputs | Portable Device Dependent | 4x DisplayPort 2.1 |

| Release Date | 2023-01-23T17:00:00.000Z | 2026-03-31T17:00:00.000Z |

The recorded data shows two GPUs with identical API support and memory type, but divergent designs across every other measured specification. The Intel Arc Pro B65 outperforms the A380M in compute, memory bandwidth, capacity, and processing unit counts, while consuming 200 W of power and requiring a desktop workstation form factor. The Intel Arc A380M offers a 35 W mobile solution with a much smaller chip and a fraction of the performance resources. The database currently contains no direct benchmark results for either product, so these specification differences represent the only quantitative basis for comparison.