Intel Arc Pro A60M vs Intel Arc Pro B370 Comparison

The Intel Arc Pro A60M and Intel Arc Pro B370 are both mobile graphics processors from Intel, but they represent two very different design philosophies and target different workload profiles. The A60M is built on the older Xe-HPG architecture using the DG2-256 chip, while the B370 is a newer design based on Xe3-LPG architecture using the Panther Lake chip. The database records no direct head-to-head benchmark results between these two parts, so the analysis below relies on their recorded specifications, architectural details, and percentile positioning to establish where each part holds an advantage.

Where Each One Wins

The Intel Arc Pro A60M is positioned for throughput-oriented tasks that benefit from a wider memory interface and a larger number of execution resources. It uses 2048 shading units, 128 texture mapping units, and 64 raster output units. Its memory subsystem is a dedicated 8 GB GDDR6 pool on a 128-bit bus, delivering 256.0 GB/s of bandwidth. This configuration gives it a clear edge in workloads that stream large data sets, such as high-resolution texture processing, multi-sample rasterization, and compute kernels that repeatedly access memory. The pixel rate of 83.20 GPixel/s and texture rate of 166.4 GTexel/s further reinforce its strength in fill-rate-bound scenarios.

The Intel Arc Pro B370, by contrast, is a low-power integrated processor built for efficiency and high clock speeds. It operates at a 25 W TDP, compared to the A60M’s 95 W TDP, and it boosts to 2400 MHz versus the A60M’s 1300 MHz. Its 1280 shading units, 40 TMUs, and 20 ROPs are fewer in count, but the higher boost clock allows it to reach 6.144 TFLOPS of FP32 compute, which is higher than the A60M’s 5.325 TFLOPS. The B370 wins in raw floating-point throughput per watt and in peak clock-driven compute. It also uses system shared memory, which means its bandwidth is system dependent, a factor that could limit it in memory-heavy tasks but also simplifies the platform.

The use-case split is therefore clear: the A60M wins in memory-bandwidth-bound professional graphics work and fill-rate-heavy rendering, while the B370 wins in compute-float throughput and power-constrained environments.

Architecture Differences

The two GPUs come from entirely different architectural generations. The A60M uses the Xe-HPG architecture, which is the foundation of Intel’s Alchemist family, specifically the Pro-Series Mobile generation. Its chip is the DG2-256, manufactured on a 6 nm process at TSMC. The die contains 11,500 million transistors on a 269 mm² area, giving a transistor density of 42.8M per mm². This is a discrete-class GPU designed for mobile workstations, with a PCIe 4.0 x16 bus interface.

The B370 uses the Xe3-LPG architecture, part of the Arc Graphics-WM (Panther Lake) generation. Its chip, Panther Lake, is manufactured on a 3 nm process at Intel’s own foundry. The transistor count and die size are listed as unknown in the database, which limits direct comparison. The B370 uses an IGP bus interface and draws all memory from the system, meaning there is no dedicated VRAM. Its predecessor is listed as HD Graphics-WM, indicating a lineage in integrated graphics rather than discrete mobile GPUs.

Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API-level feature parity is present. Both are marked as Active in production status. The A60M was released on June 5, 2023, while the B370 is dated January 26, 2026, making the B370 a much newer design.

A key architectural difference lies in the execution resource distribution. The A60M allocates 16 ray tracing cores, while the B370 has 10. The A60M also has more TMUs and ROPs, which directly feeds its higher pixel and texture rates. The B370 compensates with a much higher boost clock, 2400 MHz versus 1300 MHz, and a lower base clock of 300 MHz versus 900 MHz, suggesting a wider dynamic range for power management.

Head-to-Head Benchmarks

The database contains no recorded head-to-head benchmark entries for these two GPUs, and both have zero benchmark scores and zero wins in the winsA and winsB fields. This means the comparison must be made on specification-derived performance indicators rather than direct measured results.

The most decisive spec-level win for the B370 is in FP32 compute. It delivers 6.144 TFLOPS, which is 15.4% higher than the A60M’s 5.325 TFLOPS. This advantage comes despite the B370 having 37.5% fewer shading units (1280 vs 2048), purely because the boost clock is 84.6% higher. The FP16 throughput follows the same pattern: the B370 reaches 12.29 TFLOPS versus the A60M’s 10.65 TFLOPS, both at a 2:1 ratio.

The A60M fights back in memory and fill rates. Its 256.0 GB/s of dedicated bandwidth is a fixed figure, while the B370’s bandwidth is listed as system dependent, meaning it has no guaranteed minimum. The A60M’s pixel rate of 83.20 GPixel/s is 73.3% higher than the B370’s 48.00 GPixel/s, and its texture rate of 166.4 GTexel/s is 73.3% higher than the B370’s 96.00 GTexel/s. These margins are substantial and directly relevant to rasterization-heavy workloads.

Power consumption is another major divider. The A60M is rated at 95 W, while the B370 is rated at 25 W. The B370 delivers higher FP32 throughput while consuming 73.7% less power, which is a significant efficiency advantage. The A60M’s higher power envelope is what enables its larger memory bus and higher fill rates, but it also means the B370 is far more suitable for thermally constrained systems.

The memory clock comparison also favors the A60M in terms of dedicated resources. It runs at 2000 MHz with 16 Gbps effective data rate, whereas the B370 uses system shared memory with no fixed clock. The A60M has a 128-bit memory bus, a figure the B370 does not specify because its memory configuration is system dependent.

The Verdict

Based on the recorded data, the Intel Arc Pro A60M is the appropriate choice for professional mobile workloads that require dedicated video memory, high memory bandwidth, and strong fill rates. Its 8 GB GDDR6 pool at 256.0 GB/s, combined with 83.20 GPixel/s pixel throughput and 166.4 GTexel/s texture throughput, makes it better suited for rendering pipelines that saturate memory and raster resources. The 16 ray tracing cores also give it a structural advantage in ray-traced scenes, though no benchmark scores confirm this.

The Intel Arc Pro B370 is the better option for compute-oriented tasks in power-constrained environments. Its 6.144 TFLOPS FP32 and 12.29 TFLOPS FP16 are the highest raw compute figures between the two, and its 25 W TDP makes it far more efficient on a per-watt basis. The 2400 MHz boost clock indicates strong single-clock performance, and the 3 nm Intel process node suggests a modern manufacturing advantage. However, its system shared memory is a limiting factor for bandwidth-sensitive workloads, as the database lists its bandwidth as system dependent rather than a fixed specification.

Professionals who need predictable, dedicated memory performance should favor the A60M. Those who prioritize compute throughput and low power draw should favor the B370. Both parts share the same API support, so software compatibility is not a differentiator. The absence of direct benchmark data means the decision rests on the specification differences outlined above.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The Intel Arc Pro B370, at 6.144 TFLOPS, is 15.4% higher than the Intel Arc Pro A60M’s 5.325 TFLOPS.

Q: Does the Intel Arc Pro A60M have dedicated video memory?

A: Yes, it has 8 GB of GDDR6 memory on a 128-bit bus, delivering 256.0 GB/s of bandwidth. The B370 uses system shared memory instead.

Q: What are the power consumption figures for each GPU?

A: The Intel Arc Pro A60M is rated at 95 W, while the Intel Arc Pro B370 is rated at 25 W.

Q: Which GPU has more ray tracing cores?

A: The Intel Arc Pro A60M has 16 ray tracing cores, while the Intel Arc Pro B370 has 10.

Q: Are the APIs supported the same on both GPUs?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: What are the process nodes for each chip?

A: The Intel Arc Pro A60M uses a 6 nm TSMC process, while the Intel Arc Pro B370 uses a 3 nm Intel process.

Specification Differences

| Specification | Intel Arc Pro A60M | Intel Arc Pro B370 |

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

| Chip | DG2-256 | Panther Lake |

| Architecture | Xe-HPG | Xe3-LPG |

| Generation | Alchemist (Pro-Series Mobile) | Arc Graphics-WM (Panther Lake) |

| Process Node | 6 nm | 3 nm |

| Foundry | TSMC | Intel |

| Transistors | 11,500 million | unknown |

| Die Size | 269 mm² | unknown |

| Base Clock | 900 MHz | 300 MHz |

| Boost Clock | 1300 MHz | 2400 MHz |

| Memory Clock | 2000 MHz, 16 Gbps effective | System Shared |

| Memory Size | 8 GB | System Shared |

| Memory Type | GDDR6 | System Shared |

| Memory Bus Width | 128 bit | System Shared |

| Memory Bandwidth | 256.0 GB/s | System Dependent |

| Shading Units | 2048 | 1280 |

| TMUs | 128 | 40 |

| ROPs | 64 | 20 |

| Ray Tracing Cores | 16 | 10 |

| Pixel Rate | 83.20 GPixel/s | 48.00 GPixel/s |

| Texture Rate | 166.4 GTexel/s | 96.00 GTexel/s |

| FP32 | 5.325 TFLOPS | 6.144 TFLOPS |

| FP16 | 10.65 TFLOPS (2:1) | 12.29 TFLOPS (2:1) |

| TDP | 95 W | 25 W |

| Bus Interface | PCIe 4.0 x16 | IGP |

| Power Connectors | Not specified | None |

| Release Date | 2023-06-05 | 2026-01-26 |

| Predecessor | Not specified | HD Graphics-WM |

| Percentile vs All GPUs | 50 | 50 |