AMD Instinct MI325X vs Intel Arc 130T Mobile Comparison

Head-to-Head Benchmarks

The database records no direct head-to-head benchmark entries for the AMD Instinct MI325X against the Intel Arc 130T Mobile. Both products hold a 50th percentile position among all GPUs in the database, and each carries an average benchmark score of zero, indicating that neither has accumulated measurable performance data in the current dataset. The absence of recorded wins for either side means the comparison must be drawn entirely from the architectural and specification fields rather than from empirical test results.

The raw computational throughput figures illustrate the scale of the divide. The AMD Instinct MI325X delivers 81.72 TFLOPS of FP32 performance, while the Intel Arc 130T Mobile manages 3.942 TFLOPS in the same precision. That places the Instinct MI325X roughly 20.7 times higher in single-precision compute. In FP16 workloads, the gap narrows in relative terms but remains enormous: the Instinct MI325X sustains 81.72 TFLOPS with a 1:1 ratio to FP32, whereas the Arc 130T Mobile reaches 7.885 TFLOPS via a 2:1 ratio. The Instinct part therefore offers more than 10 times the half-precision throughput.

Texture and pixel processing tell a similar story. The Instinct MI325X posts a texture rate of 2,553.6 GTexel/s against 123.2 GTexel/s for the Arc 130T Mobile, a multiple of roughly 20.7 that mirrors the FP32 comparison. The Intel part counters with a pixel rate of 61.60 GPixel/s, while the AMD accelerator records 0 MPixel/s, a direct consequence of its lack of raster output units. The Instinct MI325X lists zero ROPs, making it unsuitable for traditional rasterized graphics output, whereas the Arc 130T Mobile carries 28 ROPs for display-oriented workloads.

Memory capacity and bandwidth are the most extreme differentiators. The Instinct MI325X pairs 256 GB of HBM3e memory on an 8192-bit bus with 6.14 TB/s of bandwidth. The Arc 130T Mobile uses system-shared memory with system-dependent bandwidth, so no fixed capacity or throughput figure exists in the database. The AMD part also runs its memory at 1500 MHz with 6 Gbps effective speed, while the Intel part shares the host system's memory clock.

The shading engine counts reinforce the hierarchy. The Instinct MI325X contains 19,456 shading units and 1,216 texture mapping units. The Arc 130T Mobile houses 896 shading units and 56 TMUs. The AMD accelerator therefore provides roughly 21.7 times the shading units and 21.7 times the TMUs of the Intel mobile part. The Arc 130T Mobile does include 7 ray tracing cores, a feature category that is listed as null for the Instinct MI325X, which suggests the accelerator omits dedicated RT hardware entirely.

Clock behavior differs substantially as well. The Instinct MI325X runs at a 1000 MHz base clock and a 2100 MHz boost clock. The Arc 130T Mobile starts at 300 MHz base and boosts to 2200 MHz. The Intel part therefore has a higher peak clock by 100 MHz, but its far smaller execution resource pool limits the practical impact of that frequency advantage.

Power draw and physical form could not contrast more sharply. The Instinct MI325X is rated at 1000 W TDP with a suggested power supply of 1400 W and mounts as an OAM module with no power connectors listed and no display outputs. The Arc 130T Mobile consumes 35 W TDP, ships as an integrated graphics processor, and offers display output that is portable-device dependent. The bus interface also separates them: PCIe 5.0 x16 for the AMD accelerator versus IGP for the Intel mobile part.

The Verdict

The recorded data points to two entirely different product categories. The AMD Instinct MI325X is an accelerator-class part oriented toward massive parallel compute, with memory capacity, bandwidth, and shading resources that dwarf the Intel Arc 130T Mobile. The Intel Arc 130T Mobile is a low-power integrated GPU intended for portable systems, with the only advantages being its 7 ray tracing cores, its 28 ROPs for pixel output, a higher boost clock of 2200 MHz versus 2100 MHz, and support for DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, all of which are marked N/A for the Instinct part.

Buyers and system designers should choose based on workload class rather than direct competition. The Instinct MI325X fits environments that need extreme FP32 or FP16 throughput, 256 GB of HBM3e memory, and 6.14 TB/s of bandwidth, all within a 1000 W power envelope. The Arc 130T Mobile suits portable devices requiring 35 W graphics, ray tracing capability, and standard graphics API support. The 50th percentile ranking for both parts in the database reflects the absence of benchmark data, not parity in capability; the specification sheet alone shows a performance gulf that no dataset could realistically bridge.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The AMD Instinct MI325X delivers 81.72 TFLOPS of FP32 performance, while the Intel Arc 130T Mobile provides 3.942 TFLOPS.

Q: How much memory does each GPU have?

A: The AMD Instinct MI325X includes 256 GB of HBM3e memory, whereas the Intel Arc 130T Mobile uses system-shared memory with no fixed capacity listed.

Q: Does the Intel Arc 130T Mobile support ray tracing?

A: Yes, it contains 7 ray tracing cores. The AMD Instinct MI325X lists no ray tracing core count in the database.

Q: What are the power requirements for each part?

A: The Instinct MI325X has a 1000 W TDP and a suggested power supply of 1400 W. The Arc 130T Mobile has a 35 W TDP with no suggested PSU listed.

Q: Which GPU has a higher boost clock?

A: The Intel Arc 130T Mobile boosts to 2200 MHz, which is 100 MHz higher than the Instinct MI325X boost clock of 2100 MHz.

Q: What graphics APIs does each GPU support?

A: The Intel Arc 130T Mobile supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The AMD Instinct MI325X lists N/A for DirectX, OpenGL, and Vulkan.

Specification Differences

The two products differ in nearly every measurable field. The Instinct MI325X uses the Aqua Vanjaram chip with CDNA 3.0 architecture, built on a 5 nm process at TSMC with 153,000 million transistors on a 1017 mm² die, yielding a transistor density of 150.4M per mm². The Arc 130T Mobile uses the Arrow Lake-H chip with Xe-LPG+ architecture, also on a 5 nm TSMC process, but its transistor count, die size, and density are listed as unknown or null.

Clocks diverge in both base and memory. The Instinct MI325X has a 1000 MHz base clock and a 2100 MHz boost clock, with memory clocked at 1500 MHz and 6 Gbps effective. The Arc 130T Mobile has a 300 MHz base clock and a 2200 MHz boost clock, with system-shared memory clock. The memory subsystem differs completely: 256 GB HBM3e on an 8192-bit bus with 6.14 TB/s bandwidth versus system-shared memory with system-dependent bandwidth.

The execution resources show a 21.7-to-1 ratio in shading units, 19,456 versus 896, and a 21.7-to-1 ratio in TMUs, 1,216 versus 56. The ROP count is zero for the Instinct MI325X and 28 for the Arc 130T Mobile. The Instinct MI325X has no ray tracing cores listed, while the Arc 130T Mobile includes 7. Pixel rate is 0 MPixel/s for the AMD part and 61.60 GPixel/s for the Intel part. Texture rate is 2,553.6 GTexel/s versus 123.2 GTexel/s.

Power and packaging are likewise opposites. The Instinct MI325X draws 1000 W with a 1400 W suggested PSU, fits an OAM Module slot, uses PCIe 5.0 x16, and has no display outputs. The Arc 130T Mobile draws 35 W, is an IGP with an IGP bus interface, and has portable-device-dependent display outputs. The production status for the Arc 130T Mobile is Active, while the Instinct MI325X has no production status recorded. Release dates also differ: the Instinct MI325X launched on October 9, 2024, and the Arc 130T Mobile on January 12, 2025.

Architecture Differences

The Instinct MI325X is built on CDNA 3.0, a compute-oriented architecture from AMD designed for data center accelerators. The Arc 130T Mobile uses Xe-LPG+, an Intel graphics architecture for integrated mobile parts. Both are fabricated on a 5 nm TSMC process, but the AMD chip is a massive 1017 mm² die with 153,000 million transistors, while the Intel chip's dimensions and transistor count are not recorded in the database.

The Instinct MI325X has no raster output units and no display output capability, reflecting its role as a compute accelerator rather than a graphics adapter. Its FP16 throughput matches its FP32 throughput at 81.72 TFLOPS, indicating a 1:1 ratio that favors raw compute consistency. The Arc 130T Mobile runs FP16 at 7.885 TFLOPS, which is double its FP32 figure, indicating a 2:1 ratio that trades precision for throughput in supported workloads.

The Arc 130T Mobile includes dedicated ray tracing hardware with 7 RT cores, a feature absent from the Instinct MI325X's specification list. The Intel part also provides a full graphics API stack with DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, while the AMD part reports no API support, consistent with a device that has no display outputs.

Memory architecture separates the two at a fundamental level. The Instinct MI325X uses dedicated HBM3e memory with a 8192-bit bus and 6.14 TB/s bandwidth, a configuration optimized for bandwidth-hungry compute workloads. The Arc 130T Mobile relies on system-shared memory with system-dependent bandwidth, a design that reduces cost and power in mobile systems at the expense of memory performance. The power envelope reflects these choices: 1000 W for the dedicated accelerator versus 35 W for the integrated part. The generation labels also differ, with the Instinct MI325X placed in the Instinct (MIx) generation and the Arc 130T Mobile in the Arc Graphics-M (Arrow Lake) generation.