AMD Instinct MI350P vs Intel Arc Graphics 4 Xe Mobile Comparison

The Verdict

The AMD Instinct MI350P and Intel Arc Graphics 4 Xe Mobile occupy completely different corners of the GPU landscape. The MI350P is a 600 W, dual-slot, PCIe 5.0 x16 accelerator with 144 GB of HBM3e memory and no display outputs, built for compute workloads where raw throughput and massive memory capacity are the priority. The Arc Graphics 4 Xe Mobile is a 25 W integrated graphics processor inside Intel's Panther Lake chip, designed for portable devices, with system-shared memory and full DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4 API support.

The data shows the MI350P delivers 36.04 TFLOPS of FP32 performance against 2.355 TFLOPS for the Arc part, a 15.3x advantage. Texture rate favors the MI350P at 1,126.4 GTexel/s versus 73.60 GTexel/s, roughly a 15.3x gap. The Arc GPU counters with a pixel rate of 36.80 GPixel/s while the MI350P has a 0 MPixel/s pixel rate, reflecting the accelerator's lack of a raster output stage. The MI350P has 8192 shading units, 512 TMUs, and 0 ROPs, while the Arc part has 512 shading units, 32 TMUs, and 16 ROPs.

Users needing a compute accelerator for large-scale server workloads should pick the MI350P. Those building or buying a portable device that needs modern graphics API support and display output should rely on the Arc Graphics 4 Xe Mobile. The MI350P does not connect to displays, offers no DirectX support, and requires a 1000 W suggested PSU. The Arc part is an IGP with no power connectors and no standalone PSU requirement.

FAQ

Q: Which GPU has more shading units?

A: The AMD Instinct MI350P has 8192 shading units, while the Intel Arc Graphics 4 Xe Mobile has 512 shading units.

Q: What memory configurations do the two GPUs use?

A: The MI350P uses 144 GB of HBM3e on an 8192-bit bus with 8.19 TB/s bandwidth. The Arc Graphics 4 Xe Mobile uses system-shared memory, with bus width and bandwidth listed as system dependent.

Q: Which GPU supports modern graphics APIs?

A: The Intel Arc Graphics 4 Xe Mobile supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The AMD Instinct MI350P lists N/A for DirectX, OpenGL, and Vulkan.

Q: What are the power requirements?

A: The MI350P has a 600 W TDP, uses a single 16-pin power connector, and has a suggested PSU of 1000 W. The Arc Graphics 4 Xe Mobile has a 25 W TDP, no power connectors, and no suggested PSU listed.

Q: Which GPU has a higher boost clock?

A: The Intel Arc Graphics 4 Xe Mobile boosts to 2300 MHz, while the AMD Instinct MI350P boosts to 2200 MHz.

Q: What is the process node for each?

A: Both use a 3 nm process node, but the MI350P is fabricated by TSMC while the Arc Graphics 4 Xe Mobile is fabricated by Intel.

Architecture Differences

The AMD Instinct MI350P is built on CDNA 4.0 architecture and uses the MI350 128CU chip. It is part of the Instinct (MIx) generation and follows the Radeon Instinct predecessor. The Intel Arc Graphics 4 Xe Mobile uses Xe3-LPG architecture on the Panther Lake chip, part of the Arc Graphics-M (Panther Lake) generation.

The MI350P is a 3 nm TSMC part with 73,000 million transistors on a 1190 mm² die, yielding a transistor density of 61.3M per mm². The Arc Graphics 4 Xe Mobile is also on a 3 nm node, but from Intel, and its transistor count and die size are listed as unknown.

The compute layouts differ sharply. The MI350P has 8192 shading units, 512 TMUs, and 0 ROPs. The Arc GPU has 512 shading units, 32 TMUs, and 16 ROPs. The MI350P reports no ray tracing cores, while the Arc part has 4. Tensor core counts are not listed for either GPU.

Memory architecture is another major split. The MI350P uses 144 GB of HBM3e with an 8192-bit bus and 8.19 TB/s bandwidth. The Arc Graphics 4 Xe Mobile uses system-shared memory with system-dependent bandwidth. The MI350P presents no display outputs, while the Arc part's display outputs are portable-device dependent.

API support also separates the two. The Arc GPU supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The MI350P lists N/A across all three API families.

Specification Differences

Clock speeds: the MI350P has a 1000 MHz base and 2200 MHz boost. The Arc Graphics 4 Xe Mobile has a 300 MHz base and 2300 MHz boost. The MI350P memory clock is 2000 MHz with 8 Gbps effective, while the Arc part uses system-shared memory.

Memory: the MI350P has 144 GB HBM3e, 8192-bit bus, and 8.19 TB/s bandwidth. The Arc GPU has system-shared memory, system-shared type, system-shared bus width, and system-dependent bandwidth.

Compute units: the MI350P has 8192 shading units, 512 TMUs, and 0 ROPs, with 0 MPixel/s pixel rate and 1,126.4 GTexel/s texture rate. The Arc GPU has 512 shading units, 32 TMUs, and 16 ROPs, with 36.80 GPixel/s pixel rate and 73.60 GTexel/s texture rate.

Performance: the MI350P delivers 36.04 TFLOPS FP32 and 36.04 TFLOPS FP16 at a 1:1 ratio. The Arc GPU delivers 2.355 TFLOPS FP32 and 4.710 TFLOPS FP16 at a 2:1 ratio.

Power and physical: the MI350P has a 600 W TDP, dual-slot width, one 16-pin connector, and a 1000 W suggested PSU. It measures 267 mm long, 111 mm high, and 40 mm wide. The Arc GPU has a 25 W TDP, IGP slot width, no power connectors, no suggested PSU, and no listed dimensions.

Interface: the MI350P uses PCIe 5.0 x16. The Arc GPU uses an IGP bus interface.

Release dates: the MI350P releases on 2026-05-06, and the Arc Graphics 4 Xe Mobile on 2026-01-26. The Arc GPU has an active production status; the MI350P's production status is not listed.

Head-to-Head Benchmarks

Direct benchmark comparisons are not available in the database, but the recorded specification data provides a clear performance picture. The MI350P's FP32 throughput of 36.04 TFLOPS is 15.3x the Arc GPU's 2.355 TFLOPS. FP16 performance shows a different ratio: the MI350P has 36.04 TFLOPS at 1:1, and the Arc part has 4.710 TFLOPS at 2:1, so the MI350P leads by 7.7x in FP16 throughput.

Texture rate follows the FP32 pattern. The MI350P processes 1,126.4 GTexel/s versus 73.60 GTexel/s for the Arc GPU, a 15.3x lead. Pixel rate is the one metric where the Arc GPU wins outright: 36.80 GPixel/s against 0 MPixel/s for the MI350P, which has no ROPs.

Memory bandwidth shows the largest single gap. The MI350P's 8.19 TB/s from 144 GB of HBM3e on an 8192-bit bus is an order of magnitude beyond the system-dependent bandwidth of the Arc GPU, which shares system memory.

The Arc GPU does hold advantages in boost clock, reaching 2300 MHz versus 2200 MHz, and in API coverage with DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4 support. The MI350P lists no API support at all.

Where Each One Wins

The AMD Instinct MI350P wins in every compute-heavy metric. Its 36.04 TFLOPS FP32 and FP16 performance suits workloads that demand massive parallel math throughput. The 144 GB HBM3e memory with 8.19 TB/s bandwidth and 8192-bit bus supports large datasets that cannot fit in the system-shared memory of an integrated GPU. The 1,126.4 GTexel/s texture rate and PCIe 5.0 x16 interface reinforce its role as a dedicated accelerator. The 600 W TDP and 1000 W suggested PSU indicate a stationary, server-oriented installation.

The Intel Arc Graphics 4 Xe Mobile wins in portability and display-oriented features. Its 25 W TDP, IGP form factor, and lack of power connectors make it suitable for mobile devices. The 36.80 GPixel/s pixel rate, 16 ROPs, and 4 ray tracing cores enable rasterization and rendering tasks that the MI350P cannot perform. DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4 support cover modern graphics APIs. The 2300 MHz boost clock is higher than the MI350P's 2200 MHz, and the 4.710 TFLOPS FP16 figure, while lower in absolute terms, comes at a fraction of the power draw.

The data indicates a clean split: the MI350P is for compute acceleration without display output, and the Arc Graphics 4 Xe Mobile is for integrated graphics in portable systems with full API support and rendering capabilities.