AMD Instinct MI300A vs Intel Arc 130V Mobile Comparison

Head-to-Head Benchmarks

The recorded data for the AMD Instinct MI300A and the Intel Arc 130V Mobile contains no overlapping benchmark scores. The MI300A has an average benchmark score of 0, and the Arc 130V Mobile also has an average benchmark score of 0. Both parts sit at the 50th percentile against all GPUs in the database, though this percentile reflects their respective positions within entirely different performance classes rather than any direct contest. With zero head-to-head benchmark entries and zero wins recorded for either side, the comparison rests entirely on their architectural and specification differences.

The MI300A delivers 61.29 TFLOPS of FP32 compute, a figure that dwarfs the Arc 130V Mobile’s 3.315 TFLOPS. That is a ratio of roughly 18.5 to 1 in raw single-precision throughput. The texture rate tells a similar story: 1,915.2 GTexel/s versus 103.6 GTexel/s, meaning the MI300A processes texture data at nearly 18.5 times the rate of the Intel part. The MI300A’s pixel rate, however, is recorded as 0 MPixel/s, as the accelerator has no display outputs and is not designed for rasterization in the traditional sense. The Arc 130V Mobile produces 51.80 GPixel/s, a figure that becomes relevant only in the context of rendering to a display.

Memory bandwidth separates these two even further. The MI300A accesses 128 GB of HBM3 across an 8192-bit bus, achieving 5.32 TB/s. The Arc 130V Mobile uses system shared memory with bandwidth described as system dependent, meaning its effective throughput varies with the host platform. The MI300A’s memory clock runs at 1300 MHz with 5.2 Gbps effective data rate. The Arc 130V Mobile’s memory clock is listed simply as system shared, with no dedicated figure.

Clock speeds show contrasting design philosophies. The MI300A has a base clock of 1000 MHz and a boost clock of 2100 MHz. The Arc 130V Mobile idles and boosts much lower, with a 300 MHz base and 1850 MHz boost. The MI300A’s boost clock is only 13.5% higher than the Arc’s, yet its FP32 output is an order of magnitude larger because it packs 14,592 shading units versus 896. The TMU counts follow suit: 912 on the MI300A versus 56 on the Arc 130V Mobile. ROPs are absent on the MI300A (0), while the Arc 130V Mobile carries 28.

The Arc 130V Mobile includes 7 ray tracing cores; the MI300A reports no ray tracing cores at all. The MI300A’s FP16 output is not listed, whereas the Arc 130V Mobile delivers 6.630 TFLOPS of FP16 through a 2:1 ratio, exactly double its FP32 number.

Where Each One Wins

The MI300A wins decisively in compute density, memory capacity, and memory bandwidth. Its 128 GB of HBM3 with 5.32 TB/s bandwidth suits workloads that load large datasets into memory and keep them resident for repeated processing, such as large model inference or scientific simulation. The 8192-bit bus width is the widest in the database for these two parts, and the 5 nm process node from TSMC packs 153,000 million transistors into a 1017 mm² die, producing a transistor density of 150.4M per mm². The MI300A’s 61.29 TFLOPS FP32 and 1,915.2 GTexel/s texture rate make it a dedicated accelerator for throughput-oriented tasks.

The Arc 130V Mobile wins in portability, available rendering features, and API support. Its 37 W TDP is a fraction of the MI300A’s 750 W. It fits as an IGP, meaning it is embedded into a processor package and requires no separate power connectors or OAM module slot. It supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, while the MI300A reports N/A for DirectX, OpenGL, and Vulkan. The Arc 130V Mobile has 28 ROPs and 7 ray tracing cores, enabling real-time graphics workloads that the MI300A cannot handle due to its lack of display outputs and zero pixel rate. The Intel part’s 51.80 GPixel/s fill rate and 103.6 GTexel/s texture rate, while far below the MI300A’s texture throughput, are sufficient for integrated graphics duties in a mobile context.

The Arc 130V Mobile also wins on process technology. Its 3 nm TSMC node is smaller than the MI300A’s 5 nm node, and its die size of 172 mm² is dramatically smaller than 1017 mm². This allows the Intel part to operate within a 37 W envelope, whereas the MI300A requires a 1150 W suggested PSU despite having no power connectors listed on the module itself.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The AMD Instinct MI300A delivers 61.29 TFLOPS of FP32, while the Intel Arc 130V Mobile delivers 3.315 TFLOPS, making the MI300A approximately 18.5 times faster in single-precision throughput.

Q: Does the Intel Arc 130V Mobile support modern graphics APIs?

A: Yes. It supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The AMD Instinct MI300A reports N/A for all three APIs.

Q: How much memory does each GPU have?

A: The MI300A has 128 GB of HBM3 memory with a 5.32 TB/s bandwidth. The Arc 130V Mobile uses system shared memory with bandwidth described as system dependent.

Q: What is the power consumption difference?

A: The MI300A has a TDP of 750 W with a suggested PSU of 1150 W. The Arc 130V Mobile has a TDP of 37 W and no suggested PSU listed.

Q: Does the MI300A have ray tracing capability?

A: No ray tracing cores are listed for the MI300A. The Arc 130V Mobile includes 7 ray tracing cores.

Q: What are the release dates?

A: The MI300A was released on 2023-12-05. The Arc 130V Mobile was released on 2024-09-23.

Specification Differences

The two parts differ across nearly every measured specification. The MI300A uses a 5 nm TSMC process, while the Arc 130V Mobile uses a 3 nm TSMC process. Transistor counts are 153,000 million for the MI300A versus unknown for the Arc 130V Mobile. Die sizes are 1017 mm² and 172 mm², respectively. Transistor density is 150.4M per mm² for the MI300A and not listed for the Arc.

Clock speeds differ: the MI300A runs at 1000 MHz base and 2100 MHz boost; the Arc 130V Mobile runs at 300 MHz base and 1850 MHz boost. Memory configurations are entirely different: the MI300A has 128 GB HBM3 with an 8192-bit bus and 5.32 TB/s bandwidth, while the Arc uses system shared memory with system dependent bandwidth. The MI300A’s memory clock is 1300 MHz (5.2 Gbps effective); the Arc’s is system shared.

Shading units number 14,592 on the MI300A versus 896 on the Arc. TMUs are 912 versus 56. ROPs are 0 versus 28. Ray tracing cores are absent on the MI300A and number 7 on the Arc. Pixel rates are 0 MPixel/s versus 51.80 GPixel/s. Texture rates are 1,915.2 GTexel/s versus 103.6 GTexel/s. FP32 output is 61.29 TFLOPS versus 3.315 TFLOPS. The Arc lists FP16 at 6.630 TFLOPS (2:1); the MI300A has no FP16 figure.

TDPs are 750 W versus 37 W. Slot widths are OAM Module versus IGP. Power connectors are none for the MI300A and not listed for the Arc. The suggested PSU is 1150 W for the MI300A and not listed for the Arc. Bus interfaces are PCIe 5.0 x16 versus IGP. Display outputs are none versus portable device dependent. API support is N/A across the board for the MI300A, while the Arc supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Production status is not listed for the MI300A and active for the Arc. Release dates are 2023-12-05 versus 2024-09-23. Predecessors are Radeon Instinct for the MI300A and HD Graphics-M for the Arc.

Architecture Differences

The MI300A is built on CDNA 3.0 architecture under the Instinct (MIx) generation, using the Aqua Vanjaram chip. The Arc 130V Mobile uses Xe2-LPG architecture under the Arc Graphics-M (Lunar Lake) generation, with the Lunar Lake chip. These are fundamentally different design targets: CDNA is optimized for compute acceleration without rasterization hardware, while Xe2-LPG is a graphics-oriented architecture with full rendering pipelines.

The MI300A’s 5 nm node with 153,000 million transistors on a 1017 mm² die indicates a massive, power-hungry accelerator. The Arc 130V Mobile’s 3 nm node with unknown transistor count on a 172 mm² die reflects an integrated solution designed for power efficiency. The MI300A’s 8192-bit memory bus and HBM3 stack provide 5.32 TB/s bandwidth, enabling data movement that the Arc’s system shared memory cannot match. The Arc’s memory architecture relies on the host system’s memory controller, making bandwidth variable.

The MI300A has no ROPs and no display outputs, confirming its role as a compute-only device. It also lacks ray tracing cores, further differentiating it from graphics-oriented parts. The Arc 130V Mobile includes 28 ROPs and 7 ray tracing cores, plus full graphics API support, positioning it for rendering tasks. The MI300A’s texture rate of 1,915.2 GTexel/s is far higher than the Arc’s 103.6 GTexel/s, but the MI300A’s zero pixel rate means it cannot output frames to a display.

The MI300A’s power delivery requires a 750 W TDP and a 1150 W suggested PSU, using an OAM Module slot with no onboard power connectors. The Arc 130V Mobile operates at 37 W as an IGP, drawing power through the host processor’s socket. The MI300A connects via PCIe 5.0 x16, while the Arc uses a direct IGP bus interface. The Arc’s production status is active; the MI300A’s is not listed. The MI300A’s predecessor is Radeon Instinct, and the Arc’s is HD Graphics-M, reflecting their separate lineage within each manufacturer’s product stack.