AMD Instinct MI355X vs Intel Iris Xe Graphics 80EU Mobile Comparison

The AMD Instinct MI355X and Intel Iris Xe Graphics 80EU Mobile occupy opposite ends of the GPU spectrum. The MI355X is a 1400 W OAM accelerator built for server racks, while the Iris Xe is a 15 W integrated processor inside laptops. The database records no shared benchmarks between them, so this analysis relies on the recorded specifications, raw compute figures, and architecture details from each entry. The data shows two devices designed for entirely different tasks, with no overlap in physical form, memory subsystem, or intended workload.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark entries for these two products. The recorded benchmark arrays for both the AMD Instinct MI355X and the Intel Iris Xe Graphics 80EU Mobile are empty, and the win counters for each side are zero. This absence of direct comparison data is itself informative: the two devices never appear in the same test suite, which reflects their divergent deployment targets.

What the data does provide is raw compute throughput. The MI355X delivers 78.64 TFLOPS of FP32 performance and 78.64 TFLOPS of FP16 performance with a 1:1 ratio. The Iris Xe delivers 1.856 TFLOPS FP32 and 3.712 TFLOPS FP16 with a 2:1 ratio. Dividing the MI355X’s FP32 figure by the Iris Xe’s FP32 figure shows the AMD part is roughly 42 times faster in single-precision work. For FP16, the MI355X is about 21 times faster than the Iris Xe’s peak FP16 rate.

Texture processing tells a similar story. The MI355X reaches 2,457.6 GTexel/s, while the Iris Xe manages 58.00 GTexel/s. That is a 42-fold advantage for the AMD accelerator. Pixel rate flips the comparison: the MI355X records 0 MPixel/s because it has no traditional raster output units and no display outputs, while the Iris Xe handles 29.00 GPixel/s. The Intel part is the only one of the two that can drive a display, and its pixel pipeline is functional for graphics output.

Clock speeds also differ sharply. The MI355X runs at a 1000 MHz base and 2400 MHz boost. The Iris Xe runs at 300 MHz base and 1450 MHz boost. The AMD part’s boost clock is 65% higher than the Intel part’s boost clock, but the real separation comes from the sheer width of the compute units: 16,384 shading units versus 640, and 1,024 TMUs versus 40.

Architecture Differences

The MI355X uses the CDNA 4.0 architecture, built on a 3 nm process at TSMC. The chip is designated MI350 256CU, and it belongs to the Instinct (MIx) generation. The Iris Xe uses Generation 12.2 architecture, built on a 10 nm process at Intel, with the chip designation Raptor Lake and the generation label HD Graphics-M (Raptor Lake). The process node gap is substantial: 3 nm versus 10 nm.

Transistor counts confirm the scale difference. The MI355X packs 185,000 million transistors on a 2380 mm² die, yielding a transistor density of 77.7M per mm². The Iris Xe entry does not list transistor count, die size, or density, so the database provides no comparable figure. The MI355X’s die is among the largest accelerators on record, while the Iris Xe is a small integrated block inside a laptop processor.

Memory systems are fundamentally different. The MI355X carries 288 GB of HBM3e on an 8192-bit bus, with 8.19 TB/s of bandwidth and a memory clock of 2000 MHz (8 Gbps effective). The Iris Xe uses System Shared memory, meaning it borrows from the host laptop’s RAM. Its bus width is also System Shared, and its bandwidth is listed as System Dependent. There is no fixed memory size, no dedicated VRAM, and no independent memory clock for the Intel part.

The compute configuration diverges completely. The MI355X has 16,384 shading units, 1,024 TMUs, and 0 ROPs. The Iris Xe has 640 shading units, 40 TMUs, and 20 ROPs. Neither product lists RT cores or tensor cores in the database, so those fields remain null for both. The MI355X’s pixel rate is 0 MPixel/s, consistent with its lack of display outputs and its role as a compute accelerator rather than a graphics card.

Power and physical design reinforce the split. The MI355X has a TDP of 1400 W, uses an OAM Module slot width, has no power connectors listed, and requires a suggested PSU of 1800 W. The Iris Xe has a TDP of 15 W, is an IGP (integrated graphics processor) with a Ring Bus interface, and lists no power connectors or PSU requirement. The MI355X measures 102 mm by 165 mm, while the Iris Xe has no recorded dimensions because it is soldered onto a motherboard.

API support tells a similar story. The MI355X lists N/A for DirectX, OpenGL, and Vulkan, confirming it is not a general-purpose graphics device. The Iris Xe supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. Display outputs for the MI355X are listed as “No outputs,” while the Iris Xe’s outputs are “Portable Device Dependent,” meaning they vary by laptop model.

Release dates differ by roughly two and a half years. The Iris Xe appeared on 2023-01-03, and the MI355X appeared on 2025-06-11. The MI355X’s predecessor is listed as Radeon Instinct, while the Iris Xe’s successor is Arc Graphics-M. Production status for the MI355X is null in the database, while the Iris Xe is marked Active.

Where Each One Wins

The MI355X wins in every category that matters for compute acceleration. Its 78.64 TFLOPS FP32 and FP16 rates are orders of magnitude above the Iris Xe’s 1.856 TFLOPS FP32 and 3.712 TFLOPS FP16. The 8.19 TB/s memory bandwidth dwarfs any System Shared configuration, and the 288 GB capacity allows working sets that an integrated GPU cannot approach. The 8192-bit bus width is a server-class feature, while the Iris Xe relies on the host’s memory controller.

The Iris Xe wins in graphics output and power efficiency. It produces 29.00 GPixel/s, which the MI355X cannot match because it records 0 MPixel/s and has no display outputs. The Iris Xe runs at 15 W, 1.07% of the MI355X’s 1400 W TDP, and it does not require an 1800 W PSU. It supports DirectX 12, OpenGL 4.6, and Vulkan 1.4, all of which are N/A for the MI355X. For any laptop task that involves rendering to a screen, the Iris Xe is the only functional option.

The texture rate comparison also shows a clear winner per side. The MI355X posts 2,457.6 GTexel/s, which is 42 times the Iris Xe’s 58.00 GTexel/s. However, the Iris Xe’s 20 ROPs and 29.00 GPixel/s mean it can actually rasterize images, while the MI355X’s 0 ROPs mean it cannot. The MI355X is built for shader and compute workloads, not for pixel generation.

Clock behavior favors the MI355X in absolute terms. Its 2400 MHz boost is 65% higher than the Iris Xe’s 1450 MHz boost, and its 1000 MHz base is more than three times the Iris Xe’s 300 MHz base. But the Iris Xe’s low clocks are a deliberate trade for power efficiency in a mobile form factor. The MI355X’s high clocks, combined with 16,384 shading units, drive its massive FP32 throughput.

The memory type also dictates use cases. HBM3e on the MI355X is designed for sustained bandwidth-hungry operations like large model inference or scientific simulation. System Shared memory on the Iris Xe is flexible but slow, and its bandwidth is System Dependent, meaning performance varies with the laptop’s RAM configuration. The MI355X has a fixed 8.19 TB/s, independent of any host system.

FAQ

Q: Which GPU has higher FP32 performance?

A: The AMD Instinct MI355X delivers 78.64 TFLOPS FP32, compared to 1.856 TFLOPS for the Intel Iris Xe Graphics 80EU Mobile. The MI355X is approximately 42 times faster in single-precision compute.

Q: Can the AMD Instinct MI355X output video to a display?

A: No. The MI355X has no display outputs and records 0 MPixel/s pixel rate. Its DirectX, OpenGL, and Vulkan support are all listed as N/A. The Intel Iris Xe, by contrast, supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, with display outputs dependent on the portable device.

Q: How do the memory systems compare?

A: The MI355X uses 288 GB of HBM3e on an 8192-bit bus with 8.19 TB/s bandwidth. The Iris Xe uses System Shared memory with System Shared bus width and System Dependent bandwidth, meaning it has no dedicated VRAM and relies on the host laptop’s RAM.

Q: What is the power consumption difference?

A: The MI355X has a TDP of 1400 W and a suggested PSU of 1800 W. The Iris Xe has a TDP of 15 W and lists no PSU requirement. The Iris Xe consumes about 1.07% of the MI355X’s power budget.

Q: Which architecture and process node does each use?

A: The MI355X uses CDNA 4.0 on a 3 nm TSMC process with the MI350 256CU chip. The Iris Xe uses Generation 12.2 on a 10 nm Intel process with the Raptor Lake chip.

Q: Does either GPU have dedicated ray tracing or tensor cores?

A: The database lists RT cores and tensor cores as null for both products. Neither entry records dedicated ray tracing hardware or tensor core counts.

The Verdict

The data points to a simple conclusion: these are not competing products. The AMD Instinct MI355X is a 1400 W, 288 GB HBM3e accelerator with 78.64 TFLOPS FP32 and no display output. It belongs in a server rack, powered by an 1800 W PSU, running compute workloads that need 8.19 TB/s of bandwidth. The Intel Iris Xe Graphics 80EU Mobile is a 15 W integrated GPU inside a laptop, with 640 shading units, System Shared memory, and full DirectX 12, OpenGL 4.6, and Vulkan 1.4 support. It exists to render a desktop, play video, and accelerate light graphics tasks on battery power.

For anyone selecting a compute accelerator for dense FP32 or FP16 math, the MI355X is the only choice in this pair. Its 16,384 shading units, 1,024 TMUs, 2400 MHz boost clock, and 8.19 TB/s memory bandwidth provide a scale of performance that the Iris Xe cannot approach. The Iris Xe’s 1.856 TFLOPS FP32 and 58.00 GTexel/s texture rate are sufficient for integrated graphics duties, but they are three orders of magnitude below the MI355X in raw throughput.

For anyone selecting a graphics solution for a laptop, the MI355X is disqualified by its lack of display outputs, N/A API support, and 0 MPixel/s pixel rate. The Iris Xe’s 29.00 GPixel/s and 20 ROPs make it a functional pixel pusher, and its 15 W TDP fits within mobile power envelopes. The MI355X cannot even initialize a screen, let alone sustain one.

The release timeline also matters. The Iris Xe launched in January 2023 and is marked Active, with a successor named Arc Graphics-M. The MI355X launched in June 2025 with no successor listed. The MI355X’s predecessor is Radeon Instinct, placing it in AMD’s dedicated accelerator line, while the Iris Xe sits in Intel’s integrated graphics family. The database shows no benchmarks for either product, so percentile rankings sit at 50 for both, but the recorded specs make the performance gulf explicit.

The verdict from the recorded data: choose the MI355X for server-side compute, choose the Iris Xe for mobile graphics. There is no middle ground, and no benchmark data exists to suggest any overlap.