Intel Arc B370 vs Intel Arc Pro B70 Comparison

Intel Arc B370 and Intel Arc Pro B70 occupy very different positions in the database. The B370 is an integrated graphics processor built into a Panther Lake mobile platform, while the Pro B70 is a large discrete workstation card built for the Battlemage Pro series. Their recorded specifications and benchmark data show almost no overlap in target use cases, performance class, or physical design. The following sections compare the two strictly on the basis of recorded measurements and listed specifications.

Head-to-Head Benchmarks

The database contains a single benchmark result for the Intel Arc B370: a 3DMark Steel Nomad DX12 score of 1184. That score places the B370 at the 5th percentile among all GPUs in the database, meaning 95% of recorded graphics cards score higher. The Intel Arc Pro B70 has no benchmark entries in the database, so its average benchmark score is recorded as 0 and no direct head-to-head comparison can be made from measured results.

The B370’s nearest rivals in the database all sit within a narrow band around its score. The ATI Mobility Radeon HD 5570 averages 1186, which is 0.2% higher than the B370. The ATI Radeon HD 5770 averages 1190, 0.5% higher. The AMD Radeon HD 7650M averages 1192, 0.7% higher. The AMD FirePro M2000 averages 1168, which is 1.4% lower than the B370. All four of these comparisons place the B370 within roughly one to two percentage points of older discrete and mobile parts, suggesting the integrated part delivers performance comparable to those early 2010s GPUs in the Steel Nomad workload.

With no benchmark results recorded for the Pro B70, the only quantitative performance comparison available is the raw specification-derived figures. The Pro B70’s FP32 throughput is listed at 22.94 TFLOPS, while the B370 is listed at 6.144 TFLOPS. That puts the Pro B70 at roughly 3.73 times the FP32 compute rate of the B370. The texture rate difference is larger: the Pro B70 lists 716.8 GTexel/s versus 96.00 GTexel/s for the B370, a factor of about 7.47. Pixel rate shows a similar gap, with the Pro B70 at 358.4 GPixel/s versus 48.00 GPixel/s for the B370, a factor of roughly 7.47 as well. These derived ratios are not benchmark scores, but they indicate the Pro B70 is designed for a much higher throughput class.

The B370’s percentile ranking at 5 confirms that its measured performance is low in absolute terms. The Pro B70’s percentile is recorded at 50, though with no measured score, that percentile cannot be tied to any specific workload result in the database. The distance between the two parts is therefore best expressed through the compute and fill-rate specifications rather than through direct performance measurements.

Where Each One Wins

The Intel Arc B370 wins in portability and integration. It is listed as an IGP with a slot width of IGP, meaning it is built into the host processor and does not occupy an expansion slot. Its TDP is 25 W, and it requires no power connectors. The memory interface is listed as System Shared, meaning it uses the host system’s memory rather than dedicated VRAM. Its display outputs are described as Portable Device Dependent, which fits an integrated part intended for laptops or compact mobile systems. The B370 is the only one of the two with a recorded benchmark score, so in the single measured workload, it is the only part with actual performance data.

The Intel Arc Pro B70 wins in every raw throughput category recorded. It lists 4096 shading units against 1280 for the B370, 256 TMUs against 40, 128 ROPs against 20, and 32 ray tracing cores against 10. Its FP32 throughput of 22.94 TFLOPS is nearly four times the B370’s 6.144 TFLOPS. The Pro B70 also lists dedicated GDDR6 memory totaling 32 GB on a 256 bit bus, with bandwidth of 608.0 GB/s, while the B370’s memory bandwidth is listed as System Dependent. The Pro B70’s memory clock is 2375 MHz with 19 Gbps effective transfer, while the B370’s memory clock is listed as System Shared. In any memory-bound or compute-bound scenario, the Pro B70’s dedicated memory subsystem and higher shader count give it a clear advantage.

The Pro B70 also wins on display connectivity. It lists 1x HDMI 2.1a and 3x DisplayPort 2.1 outputs, whereas the B370’s outputs are dependent on the portable device it is integrated into. The Pro B70 uses a PCIe 5.0 x16 bus interface, while the B370 uses IGP. The Pro B70 is a dual-slot card measuring 267 mm in length, 110 mm in height, and 39 mm in width, with a 1x 8-pin power connector and a suggested 550 W power supply. None of those physical attributes apply to the B370, which is an integrated part with no power connectors and no listed dimensions.

The B370 wins on power draw. Its 25 W TDP is far below the Pro B70’s 230 W TDP. The B370 also has a lower base clock of 300 MHz versus 2280 MHz for the Pro B70, and a lower boost clock of 2400 MHz versus 2800 MHz. Lower clocks and lower power draw make the B370 suitable for systems where thermal and power budgets are tight, while the Pro B70 is built for sustained high throughput in a desktop or workstation chassis.

Architecture Differences

The two GPUs come from different Intel graphics architectures and are manufactured on different process nodes. The Intel Arc B370 uses the Xe3-LPG architecture and is built on a 3 nm process at Intel’s own foundry. Its chip is listed as Panther Lake, and its generation is Arc Graphics-M (Panther Lake). The Intel Arc Pro B70 uses the Xe2-HPG architecture and is built on a 5 nm process at TSMC. Its chip is listed as BMG-G31, and its generation is Battlemage (Pro Series).

The process node difference is notable. The B370’s 3 nm node is smaller than the Pro B70’s 5 nm node, which typically allows for denser transistor packing and lower power per transistor, though the database does not list transistor counts for either part. The die size is unknown for the B370, while the Pro B70 lists a die size of 368 mm². The foundry differs as well: Intel fabricates the B370, while TSMC fabricates the Pro B70.

Architecture generation also differs. The B370’s Xe3-LPG is described as part of the Arc Graphics-M line, which aligns with its integrated, low-power positioning. The Pro B70’s Xe2-HPG is part of the Battlemage Pro Series, which aligns with its high-end discrete workstation positioning. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so the API feature sets are recorded as identical.

The B370’s base clock is 300 MHz, which is very low for a GPU, but its boost clock reaches 2400 MHz. The Pro B70’s base clock is 2280 MHz and its boost clock is 2800 MHz. The B370’s low base clock likely reflects its integrated nature and power management strategy, while the Pro B70’s high base clock reflects a discrete card designed to run at sustained high frequencies.

The memory architecture differs fundamentally. The B370 uses System Shared memory for both size and type, with a bus width listed as System Shared and bandwidth listed as System Dependent. The Pro B70 uses 32 GB of GDDR6 on a 256 bit bus with 608.0 GB/s bandwidth and a memory clock of 2375 MHz with 19 Gbps effective transfer. This means the Pro B70 has its own dedicated memory pool, while the B370 depends on the host system’s memory configuration.

The core counts also show a large architectural gap. The B370 has 1280 shading units, 40 TMUs, 20 ROPs, and 10 RT cores. The Pro B70 has 4096 shading units, 256 TMUs, 128 ROPs, and 32 RT cores. The Pro B70 has exactly 3.2 times the shading units, 6.4 times the TMUs, 6.4 times the ROPs, and 3.2 times the RT cores of the B370. The TMU and ROP ratios match each other but differ from the shader and RT core ratios, which suggests the two architectures allocate resources differently between compute, texture, and pixel processing.

Specification Differences

The two cards differ in nearly every recorded specification category. The B370 is an IGP with slot width IGP, no power connectors, no suggested PSU, and no dimensions. The Pro B70 is a dual-slot card with a 1x 8-pin power connector, a suggested 550 W PSU, and dimensions of 267 mm length, 110 mm height, and 39 mm width.

Process node: the B370 is on 3 nm from Intel, the Pro B70 is on 5 nm from TSMC. Die size is unknown for the B370 and 368 mm² for the Pro B70. Transistor count is unknown for both.

Clocks: the B370 lists base 300 MHz, boost 2400 MHz, and memory as System Shared. The Pro B70 lists base 2280 MHz, boost 2800 MHz, and memory clock 2375 MHz with 19 Gbps effective.

Memory: the B370 has System Shared size, type, bus width, and System Dependent bandwidth. The Pro B70 has 32 GB GDDR6, 256 bit bus, and 608.0 GB/s bandwidth.

Compute: the B370 has 1280 shading units, 40 TMUs, 20 ROPs, 10 RT cores, 48.00 GPixel/s pixel rate, 96.00 GTexel/s texture rate, 6.144 TFLOPS FP32, and 12.29 TFLOPS FP16 (2:1). The Pro B70 has 4096 shading units, 256 TMUs, 128 ROPs, 32 RT cores, 358.4 GPixel/s pixel rate, 716.8 GTexel/s texture rate, 22.94 TFLOPS FP32, and 45.88 TFLOPS FP16 (2:1).

Power: the B370 has a TDP of 25 W. The Pro B70 has a TDP of 230 W. The B370 has no power connectors, while the Pro B70 uses 1x 8-pin.

Bus interface: the B370 uses IGP, the Pro B70 uses PCIe 5.0 x16.

Display outputs: the B370 lists Portable Device Dependent, the Pro B70 lists 1x HDMI 2.1a and 3x DisplayPort 2.1.

Release dates: the B370 is dated 2026-01-26, the Pro B70 is dated 2026-03-25. Production status is listed as Active for the B370 and null for the Pro B70. The Pro B70 has a launch MSRP of 949 USD. The B370 has no launch MSRP listed.

Benchmark data: the B370 has one recorded score of 1184 in 3DMark Steel Nomad DX12, with an average benchmark score of 1184 and a percentile of 5. The Pro B70 has no recorded benchmark scores, an average benchmark score of 0, and no nearest rivals listed. The B370’s nearest rivals are the ATI Mobility Radeon HD 5570 at 1186, the ATI Radeon HD 5770 at 1190, the AMD Radeon HD 7650M at 1192, and the AMD FirePro M2000 at 1168. The Pro B70 has no nearest rivals in the database.

FAQ

Q: Which GPU has a higher recorded benchmark score?

A: The Intel Arc B370 has a recorded 3DMark Steel Nomad DX12 score of 1184. The Intel Arc Pro B70 has no recorded benchmark scores in the database, so its average benchmark score is 0.

Q: How much faster is the Pro B70 in FP32 compute?

A: The Pro B70 lists FP32 throughput of 22.94 TFLOPS, while the B370 lists 6.144 TFLOPS. The Pro B70’s FP32 rate is approximately 3.73 times the B370’s.

Q: What memory configurations do the two GPUs use?

A: The B370 uses System Shared memory with System Dependent bandwidth. The Pro B70 uses 32 GB of GDDR6 on a 256 bit bus with 608.0 GB/s bandwidth and a memory clock of 2375 MHz with 19 Gbps effective transfer.

Q: How do their power requirements compare?

A: The B370 has a TDP of 25 W and no power connectors. The Pro B70 has a TDP of 230 W and requires a 1x 8-pin power connector, with a suggested 550 W power supply.

Q: What are the physical dimensions of each card?

A: The B370 is an IGP with no listed dimensions. The Pro B70 is a dual-slot card measuring 267 mm in length, 110 mm in height, and 39 mm in width.

Q: Which GPU has more shading units and ray tracing cores?

A: The Pro B70 has 4096 shading units and 32 RT cores. The B370 has 1280 shading units and 10 RT cores. The Pro B70 has 3.2 times the shading units and 3.2 times the RT cores of the B370.