Intel Arc A380E vs NVIDIA GB10 Comparison

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark runs for the Intel Arc A380E and the NVIDIA GB10. The Intel Arc A380E has no recorded benchmark scores, while the NVIDIA GB10 has two: a Geekbench OpenCL score of 120137 and a Geekbench Vulkan score of 114648. Its average benchmark score is 117393, placing it in the 95th percentile among all GPUs.

The NVIDIA GB10's nearest rivals in the database provide the context for interpreting its scores. It sits 0.3% ahead of the NVIDIA RTX 4000 SFF Ada Generation (117088), 1.3% behind the AMD Radeon PRO W7700 (118976), 2.6% ahead of the NVIDIA Tesla V100 SXM2 16 GB (114395), and 3% ahead of the NVIDIA RTX A5500 Mobile (113944). These deltas are all small, indicating the GB10 lands in a tightly contested performance cluster.

For the Intel Arc A380E, the absence of benchmark entries means no percentile or rival comparisons exist. The data simply does not include measurable performance for this part. Any direct numerical comparison between the two would be fabricated, so the analysis must rely on architectural and specification differences instead.

Where Each One Wins

The NVIDIA GB10 holds every measurable performance advantage in the recorded data. Its 29.71 TFLOPS FP32 throughput is roughly 7.3 times the Intel Arc A380E's 4.096 TFLOPS. In FP16, the GB10 delivers 29.71 TFLOPS at a 1:1 ratio, while the Arc A380E produces 8.192 TFLOPS at a 2:1 ratio. The GB10's texture rate of 928.5 GTexel/s dwarfs the Arc A380E's 128.0 GTexel/s, and its pixel rate of 116.1 GPixel/s nearly doubles the Arc A380E's 64.00 GPixel/s.

Memory capacity is another decisive split. The GB10 carries 128 GB of LPDDR5X on a 256-bit bus, yielding 273.2 GB/s of bandwidth. The Arc A380E has 6 GB of GDDR6 on a 96-bit bus, producing 186.0 GB/s. The GB10's bandwidth advantage is 1.5 times, but its capacity advantage is over 21 times. That capacity gap changes the class of workloads each can address.

The GB10 also leads in compute resources. It has 6144 shading units, 384 TMUs, 48 ROPs, 48 RT cores, and 384 tensor cores. The Arc A380E has 1024 shading units, 64 TMUs, 32 ROPs, and 8 RT cores, with no tensor cores listed. The GB10's boost clock reaches 2418 MHz versus the Arc A380E's fixed 2000 MHz (base and boost are identical).

The Intel Arc A380E does hold advantages in a few areas, though they are narrower. Its memory runs at 1937 MHz (15.5 Gbps effective) versus the GB10's 1067 MHz (8.5 Gbps effective). The Arc A380E also supports a full display suite with 4x DisplayPort 2.0 outputs, while the GB10 lists only 1x HDMI. The Arc A380E carries API support for DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4; the GB10 lists N/A for all three.

The Verdict

The recorded data points to two different products for two different jobs. The NVIDIA GB10 is a high-performance compute device with substantial memory and processing resources. Its 95th percentile standing and average benchmark score of 117393 place it among serious workstation and server parts, slightly ahead of the RTX 4000 SFF Ada Generation and just behind the Radeon PRO W7700.

The Intel Arc A380E is a low-power graphics card with a 75 W TDP, no power connectors, and a suggested PSU of 250 W. Its single-slot, 254 mm length, and 4x DisplayPort 2.0 outputs suggest a display-oriented role. The GB10, by contrast, is an integrated graphics processor (IGP) with a 140 W TDP and a suggested PSU of 300 W.

Buyers needing raw compute, large memory capacity, and tensor core acceleration should look at the NVIDIA GB10. The data shows it delivers over 7 times the FP32 throughput, 21 times the memory capacity, and 384 tensor cores. Buyers needing a compact, low-power graphics card with modern display outputs and broad graphics API support should consider the Intel Arc A380E. The two do not compete in the same segment.

FAQ

Q: Which GPU has higher FP32 performance?

A: The NVIDIA GB10, with 29.71 TFLOPS versus the Intel Arc A380E's 4.096 TFLOPS.

Q: How much memory does each GPU have?

A: The NVIDIA GB10 has 128 GB of LPDDR5X, while the Intel Arc A380E has 6 GB of GDDR6.

Q: Which GPU has better benchmark results in the database?

A: Only the NVIDIA GB10 has recorded benchmarks: a Geekbench OpenCL score of 120137 and a Geekbench Vulkan score of 114648, averaging 117393. The Intel Arc A380E has no benchmark entries.

Q: What is the memory bandwidth difference?

A: The NVIDIA GB10 delivers 273.2 GB/s over a 256-bit bus, while the Intel Arc A380E delivers 186.0 GB/s over a 96-bit bus.

Q: Which GPU supports ray tracing?

A: Both list ray tracing hardware: the Intel Arc A380E has 8 RT cores, and the NVIDIA GB10 has 48 RT cores.

Q: What display outputs does each GPU have?

A: The Intel Arc A380E has 4x DisplayPort 2.0 outputs, while the NVIDIA GB10 has a single HDMI output.

Architecture Differences

The Intel Arc A380E uses the DG2-128 chip built on TSMC's 6 nm process, with an Xe-HPG architecture from the Alchemist generation. It contains 7,200 million transistors on a 157 mm² die, giving a transistor density of 45.9M per mm². The NVIDIA GB10 uses the GB20B chip on TSMC's 5 nm process, with a Blackwell 2.0 architecture from the Server Blackwell generation. Its die size is 382 mm², and its transistor count is listed as unknown.

The A380E is an end-of-life product released in March 2024, succeeding Xe Graphics and preceding Battlemage. The GB10 is an active product released in October 2025, succeeding Server Hopper and preceding Server Rubin. The GB10's architecture is server-oriented, while the A380E's is client graphics-oriented.

The A380E supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The GB10 lists N/A for DirectX, OpenGL, and Vulkan, indicating it is not designed around conventional graphics APIs. The A380E has no tensor cores listed, while the GB10 has 384. The A380E uses a 2:1 FP16 ratio, while the GB10 uses a 1:1 ratio, meaning the GB10 treats FP16 and FP32 workloads with equal throughput.

The GB10's memory type is LPDDR5X, a low-power DRAM typically integrated on-package, fitting its IGP slot designation. The A380E uses discrete GDDR6. The GB10's bus interface is PCIe 5.0 x16, while the A380E uses PCIe 4.0 x8.

Specification Differences

| Specification | Intel Arc A380E | NVIDIA GB10 |

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

| Process node | 6 nm | 5 nm |

| Die size | 157 mm² | 382 mm² |

| Transistors | 7,200 million | unknown |

| Base clock | 2000 MHz | 1665 MHz |

| Boost clock | 2000 MHz | 2418 MHz |

| Memory clock | 1937 MHz (15.5 Gbps effective) | 1067 MHz (8.5 Gbps effective) |

| Memory size | 6 GB GDDR6 | 128 GB LPDDR5X |

| Memory bus | 96 bit | 256 bit |

| Memory bandwidth | 186.0 GB/s | 273.2 GB/s |

| Shading units | 1024 | 6144 |

| TMUs | 64 | 384 |

| ROPs | 32 | 48 |

| RT cores | 8 | 48 |

| Tensor cores | none listed | 384 |

| Pixel rate | 64.00 GPixel/s | 116.1 GPixel/s |

| Texture rate | 128.0 GTexel/s | 928.5 GTexel/s |

| FP32 | 4.096 TFLOPS | 29.71 TFLOPS |

| FP16 | 8.192 TFLOPS (2:1) | 29.71 TFLOPS (1:1) |

| TDP | 75 W | 140 W |

| Slot width | Single-slot | IGP |

| Power connectors | None | None |

| Suggested PSU | 250 W | 300 W |

| Bus interface | PCIe 4.0 x8 | PCIe 5.0 x16 |

| Display outputs | 4x DisplayPort 2.0 | 1x HDMI |

| Dimensions | 254 mm x 127 mm x 20 mm | 150 mm x 51 mm x 150 mm |

| Production status | End-of-life | Active |

| Release date | 2024-03-31 | 2025-10-14 |

| API support | DirectX 12 Ultimate, OpenGL 4.6, Vulkan 1.4 | N/A for all |

| Percentile vs all GPUs | 50 | 95 |

| Average benchmark score | 0 | 117393 |

The GB10's launch MSRP is 3,999 USD. The A380E has no launch MSRP listed. The GB10's 382 mm² die is 2.4 times larger than the A380E's 157 mm². The A380E's 2000 MHz base and boost clocks are identical, while the GB10 boosts from 1665 MHz to 2418 MHz, a 45% increase. Neither card requires external power connectors, but the GB10's 140 W TDP nearly doubles the A380E's 75 W TDP.