AMD Instinct MI350X vs Intel Arc Graphics 112EU Mobile Comparison

# Head-to-Head Benchmarks

The database records no direct head-to-head benchmark matches for the AMD Instinct MI350X and the Intel Arc Graphics 112EU Mobile. Both entries show a percentile vs all GPUs score of 50, placing them at the midpoint of the recorded performance distribution. However, the raw computational metrics tell a stark story of scale and purpose. The MI350X delivers 72.09 TFLOPS of FP32 compute, while the Arc 112EU Mobile produces 3.942 TFLOPS. That is an 18.3x difference in raw floating-point throughput, a gap that reflects entirely different design targets.

For FP16 workloads, the MI350X holds at 72.09 TFLOPS with a 1:1 ratio, meaning it does not gain a throughput advantage from reduced precision. The Intel part, using a 2:1 ratio, reaches 7.885 TFLOPS, which is still 9.1x lower than the AMD accelerator. The texture rate follows the same pattern: the MI350X sustains 2,252.8 GTexel/s against 123.2 GTexel/s for the Arc part, a difference of roughly 18.3x. Pixel rate is the one metric where the Intel GPU posts a finite number, 52.80 GPixel/s, while the MI350X shows 0 MPixel/s because it has no raster output units and no display outputs.

The absence of shared benchmark scores means there is no measured game or compute workload where these two parts directly compete. The data instead shows two products occupying opposite ends of the hardware spectrum, one designed for massive parallel throughput and the other for integrated graphics duties in mobile systems.

# Architecture Differences

The MI350X uses the CDNA 4.0 architecture built on a 3 nm process at TSMC. It packs 185,000 million transistors onto a 2380 mm² die, yielding a transistor density of 77.7M per mm². The chip is labeled MI350 256CU and belongs to the Instinct (MIx) generation. Its memory subsystem consists of 288 GB of HBM3e on an 8192-bit bus, delivering 8.19 TB/s of bandwidth. The memory clock is listed at 2000 MHz with 8 Gbps effective data rate. Base clock is 1000 MHz, boost clock is 2200 MHz. The GPU has 16,384 shading units and 1,024 texture mapping units, with zero ROPs and no pixel output capability.

The Intel Arc Graphics 112EU Mobile is built on the Xe-LPG architecture, fabricated on Intel's 10 nm process. It is part of the Arc Graphics-M (Meteor Lake) generation. The chip has 896 shading units, 56 TMUs, and 24 ROPs. Its base clock is 300 MHz and boost clock is 2200 MHz. Memory is system-shared, meaning the GPU uses the host system's RAM for both capacity and bandwidth, which the database lists as "System Dependent." This is fundamentally different from the dedicated 288 GB HBM3e pool on the MI350X.

The MI350X uses a PCIe 5.0 x16 bus interface, while the Intel part uses a Ring Bus. The MI350X is an OAM Module with no power connectors and a suggested PSU of 1400 W. Its TDP is 1000 W. The Arc part is an IGP with a 65 W TDP and no suggested PSU listed. The AMD accelerator has no display outputs and no supported APIs (DirectX, OpenGL, Vulkan all listed as N/A). The Intel GPU supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, and its display outputs are "Portable Device Dependent."

Release dates differ by roughly 18 months: the Arc 112EU Mobile launched on 2023-12-13, while the MI350X launched on 2025-06-11. The MI350X's predecessor is Radeon Instinct; the Arc part's predecessor is HD Graphics-M. Neither has a successor listed in the database.

# The Verdict

The data indicates these two GPUs are not competitors in any practical sense. The MI350X is a 1000 W accelerator with 288 GB of dedicated HBM3e memory and 72.09 TFLOPS of FP32 compute, designed for training and inference workloads that require massive memory capacity and bandwidth. The Arc 112EU Mobile is a 65 W integrated GPU with system-shared memory, 3.942 TFLOPS of FP32 compute, and full API support for DirectX 12, OpenGL 4.6, and Vulkan 1.4, intended for laptops and compact mobile devices.

Who should pick which? The recorded data supports a clear split: any workload that can use 8.19 TB/s of memory bandwidth and 72.09 TFLOPS of compute belongs on the MI350X. Any workload that needs a display output, API compatibility, and a 65 W power envelope belongs on the Arc part. The MI350X has no display outputs and no API support, so it cannot drive a monitor or run consumer graphics software. The Arc part cannot approach the MI350X's memory capacity or compute throughput, but it does not need to, as its role is entirely different.

# Specification Differences

| Specification | AMD Instinct MI350X | Intel Arc Graphics 112EU Mobile |

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

| Architecture | CDNA 4.0 | Xe-LPG |

| Process Node | 3 nm | 10 nm |

| Foundry | TSMC | Intel |

| Transistors | 185,000 million | Not listed |

| Die Size | 2380 mm² | Not listed |

| Transistor Density | 77.7M / mm² | Not listed |

| Base Clock | 1000 MHz | 300 MHz |

| Boost Clock | 2200 MHz | 2200 MHz |

| Memory Size | 288 GB | System Shared |

| Memory Type | HBM3e | System Shared |

| Memory Bus Width | 8192 bit | System Shared |

| Memory Bandwidth | 8.19 TB/s | System Dependent |

| Shading Units | 16,384 | 896 |

| TMUs | 1,024 | 56 |

| ROPs | 0 | 24 |

| Pixel Rate | 0 MPixel/s | 52.80 GPixel/s |

| Texture Rate | 2,252.8 GTexel/s | 123.2 GTexel/s |

| FP32 | 72.09 TFLOPS | 3.942 TFLOPS |

| FP16 | 72.09 TFLOPS (1:1) | 7.885 TFLOPS (2:1) |

| TDP | 1000 W | 65 W |

| Slot Width | OAM Module | IGP |

| Power Connectors | None | Not listed |

| Suggested PSU | 1400 W | Not listed |

| Bus Interface | PCIe 5.0 x16 | Ring Bus |

| Display Outputs | No outputs | Portable Device Dependent |

| DirectX | N/A | 12 (12_1) |

| OpenGL | N/A | 4.6 |

| Vulkan | N/A | 1.4 |

| Release Date | 2025-06-11 | 2023-12-13 |

| Predecessor | Radeon Instinct | HD Graphics-M |

# FAQ

Q: Which GPU has higher FP32 compute performance?

A: The AMD Instinct MI350X delivers 72.09 TFLOPS, while the Intel Arc 112EU Mobile delivers 3.942 TFLOPS. The MI350X is about 18.3x faster in FP32.

Q: Can the MI350X output video to a display?

A: No. The database lists "No outputs" for display outputs, and its DirectX, OpenGL, and Vulkan APIs are all N/A. The Intel Arc part supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4 with portable device dependent display outputs.

Q: How much memory does each GPU have?

A: The MI350X has 288 GB of HBM3e on an 8192-bit bus. The Intel Arc part uses system-shared memory, meaning capacity and bandwidth depend on the host system.

Q: What is the memory bandwidth difference?

A: The MI350X provides 8.19 TB/s of dedicated bandwidth. The Arc 112EU Mobile's bandwidth is listed as "System Dependent," so no fixed figure is recorded.

Q: What are the power requirements?

A: The MI350X has a 1000 W TDP and a suggested PSU of 1400 W. The Arc part has a 65 W TDP and no suggested PSU listed.

Q: When did each product launch?

A: The Intel Arc 112EU Mobile launched on 2023-12-13. The AMD Instinct MI350X launched on 2025-06-11.

# Where Each One Wins

The AMD Instinct MI350X wins decisively in every computational throughput category recorded in the database. FP32 compute is 72.09 TFLOPS versus 3.942 TFLOPS. FP16 compute is 72.09 TFLOPS versus 7.885 TFLOPS. Texture rate is 2,252.8 GTexel/s versus 123.2 GTexel/s. Memory bandwidth is 8.19 TB/s versus system-dependent. Memory capacity is 288 GB versus system-shared. It also has far more shading units, 16,384 versus 896, and more TMUs, 1,024 versus 56. The MI350X uses a 3 nm process with 185,000 million transistors on a 2380 mm² die, compared to Intel's 10 nm node with no transistor or die size data recorded.

The Intel Arc Graphics 112EU Mobile wins in areas that matter for consumer and mobile use. It has 24 ROPs and a pixel rate of 52.80 GPixel/s, while the MI350X has zero ROPs and 0 MPixel/s. It supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, while the MI350X lists N/A for all three APIs. It has display outputs, described as portable device dependent, whereas the MI350X has none. Its TDP is 65 W compared to 1000 W, and it has a lower base clock of 300 MHz versus 1000 MHz, which aligns with its power profile. The Arc part also uses a Ring Bus interface, while the MI350X uses PCIe 5.0 x16. It launched earlier, on 2023-12-13, versus 2025-06-11 for the MI350X.

The use cases follow directly from these wins. The MI350X is for compute-heavy workloads that need massive memory capacity and bandwidth, high FP32 and FP16 throughput, and do not require display output or graphics API support. The Arc 112EU Mobile is for mobile systems that need a display output, API compatibility for consumer graphics, and low power consumption. Neither part can substitute for the other in its respective role, based on the recorded specifications.