AMD Ryzen AI 9 HX 375 vs Intel Arc G3 EXTREME Comparison

Head-to-Head Benchmarks

The benchmark database records 15 direct comparisons between the AMD Ryzen AI 9 HX 375 and the Intel Arc G3 EXTREME. The AMD part wins 9 of those tests, while the Intel part wins 6. The overall picture is clear: the AMD processor holds a substantial lead in most compute-heavy workloads, but the Intel chip counters with notable wins in specific math and single-threaded tasks.

Starting with the rendering benchmarks, the AMD Ryzen AI 9 HX 375 dominates. In Cinebench R15 multi-core, the AMD scores 3334 against Intel's 2192, a delta of 52.1%. The Cinebench R23 multi-core test tells a similar story: AMD at 21812 versus Intel at 14655, a 48.8% advantage. Single-core rendering also favors AMD, though by a smaller margin. In Cinebench R15 single-core, the scores are 301 versus 267, a 12.7% lead. In Cinebench R23 single-core, AMD posts 1988 against Intel's 1862, a 6.8% edge. These results indicate that the AMD architecture delivers higher throughput in both lightly threaded and heavily threaded rendering tasks.

The PassMark suite reveals a more mixed landscape. The AMD Ryzen AI 9 HX 375 wins several integer and data-processing tests decisively. In integer math, AMD scores 121754 versus Intel's 71164, a massive 71.1% advantage. Data compression also favors AMD heavily: 404918 versus 307094, a 31.9% lead. Extended instructions go to AMD at 29269 versus 24017, a 21.9% gap. Random string sorting also lands in AMD's favor: 44552 versus 37331, a 19.3% margin. The multithread test shows AMD ahead at 32916 versus 30659, a 7.4% win.

The Intel Arc G3 EXTREME, however, takes the floating-point and physics workloads. In floating-point math, Intel scores 86929 against AMD's 75153, a 13.5% advantage. The physics test shows Intel at 2515 versus AMD's 1819, a 27.7% lead. Prime number finding is another strong Intel result: 248 versus 122, a 50.8% margin. Data encryption also goes to Intel: 23881 versus 20802, a 12.9% edge. The single-thread PassMark tests show Intel ahead at 4293 versus AMD's 3867, a 9.9% difference. This pattern suggests Intel's cores handle certain scalar and floating-point operations more efficiently, while AMD's larger thread count and cache configuration drive its wins in integer-heavy and compression workloads.

Where Each One Wins

Looking at the benchmark distribution, the AMD Ryzen AI 9 HX 375 is the clear choice for multi-core rendering, integer math, data compression, and general multithreaded productivity. The Cinebench R23 multi-core score of 21812 represents a performance tier that the Intel part cannot match, sitting 48.8% behind. Similarly, the integer math score of 121754 is more than 70% higher than Intel's, which points to strong performance in code compilation, spreadsheet calculations, and other integer-heavy tasks. The data compression score of 404918 indicates solid performance in file archiving and database workloads.

The Intel Arc G3 EXTREME wins in areas that rely on floating-point throughput, physics simulation, and single-threaded execution. The floating-point math score of 86929 is 13.5% ahead of AMD, which benefits scientific computing, financial modeling, and any workload that uses heavy FPU instructions. The physics score of 2515 is 27.7% ahead, suggesting better performance in physics-based simulation and gaming physics. The prime number finding score of 248 is more than double AMD's 122, which indicates a strong integer division unit or specialized instruction path. The single-thread PassMark score of 4293 versus 3867 shows that Intel's fastest core has a 9.9% edge in lightly threaded legacy applications.

The multithread PassMark score, where AMD wins 32916 to 30659, is close at 7.4%. This suggests that in mixed workloads that combine integer, floating-point, and memory operations, the two processors are more comparable than the individual test deltas might imply. The AMD part leverages its 24 threads to pull ahead, while the Intel part relies on fewer but more efficient cores.

Architecture Differences

The two processors come from different design philosophies and manufacturing processes. The AMD Ryzen AI 9 HX 375 uses the Zen 5 architecture on a 4 nm TSMC process node. It has 12 cores and 24 threads, with a base clock of 2.00 GHz and a boost clock of 5.10 GHz. The thermal design power is 28 W. The chip uses the AMD Socket FP8 and belongs to the Strix Point codename family, part of the Ryzen AI 300 generation that mixes Zen 5 and Zen 5c cores. The die size is 233 mm².

The Intel Arc G3 EXTREME uses the Panther Lake codename on a 3 nm Intel process node. It has 14 cores and 14 threads, meaning no hyperthreading. The base clock is 1.90 GHz and the boost clock is 4.70 GHz. The thermal design power is 25 W. It uses the Intel BGA 2540 socket. The generation is listed as Arc G3 (Panther Lake). No die size is recorded in the database.

Cache configurations differ significantly. The AMD part has 80 KB of L1 per core, 1 MB of L2 per core, and 16 MB of L3 cache. The Intel part has 192 KB of L1 per core, 2.5 MB of L2 per core, and 18 MB of shared L3 cache. The Intel cache hierarchy provides more per-core L1 and L2, which may explain its single-thread and physics wins. The AMD L3 is smaller at 16 MB, but the overall cache-to-thread ratio is different given the 24 threads versus 14 threads.

Memory support also differs. The AMD Ryzen AI 9 HX 375 supports DDR5 and LPDDR5X memory over a dual-channel bus, with a recorded memory bandwidth of 89.6 GB/s. The Intel Arc G3 EXTREME supports only LPDDR5X, also dual-channel, but with a higher recorded memory bandwidth of 136.5 GB/s. The Intel part has a 52.4% higher memory bandwidth based on the recorded figures, which may contribute to its floating-point and physics performance.

PCIe connectivity differs as well. The AMD chip provides Gen 4 with 16 lanes (CPU only), while the Intel chip provides Gen 5 with 4 lanes (CPU only). The AMD part offers more lanes for expansion, while the Intel part uses a newer PCIe generation but with fewer lanes. Integrated graphics differ: AMD uses the Radeon 890M, while Intel uses the Arc B390. Both are mobile segments, and both are in active production.

The AMD part was released on 2024-06-30, while the Intel part has a release date of 2026-05-27. The AMD part has a part number of 100-000001682, and the Intel part has part number SA4R3. Neither processor has a recorded launch MSRP, and neither has an unlocked multiplier.

FAQ

Q: Which processor has more cores and threads?

A: The Intel Arc G3 EXTREME has 14 cores but only 14 threads, meaning no simultaneous multithreading. The AMD Ryzen AI 9 HX 375 has 12 cores and 24 threads, providing twice the thread count.

Q: How do the two compare in Cinebench R23 multi-core?

A: The AMD Ryzen AI 9 HX 375 scores 21812, which is 48.8% higher than the Intel Arc G3 EXTREME's 14655. This is the largest rendering gap in the database between these two processors.

Q: In which tests does the Intel Arc G3 EXTREME outperform AMD?

A: The Intel part wins in floating-point math (86929 versus 75153), physics (2515 versus 1819), prime number finding (248 versus 122), data encryption (23881 versus 20802), and PassMark single-thread (4293 versus 3867).

Q: What is the memory bandwidth difference?

A: The Intel Arc G3 EXTREME has a recorded memory bandwidth of 136.5 GB/s, while the AMD Ryzen AI 9 HX 375 has 89.6 GB/s. The Intel part also supports only LPDDR5X, while AMD supports both DDR5 and LPDDR5X.

Q: Which processor has a higher boost clock?

A: The AMD Ryzen AI 9 HX 375 has a boost clock of 5.10 GHz, which is higher than the Intel Arc G3 EXTREME's 4.70 GHz.

Q: How do the average benchmark scores compare?

A: The AMD Ryzen AI 9 HX 375 has an average benchmark score of 46030, while the Intel Arc G3 EXTREME has an average of 36699. The AMD part sits in the 89th percentile of all CPUs, while the Intel part sits in the 85th percentile.

The Verdict

The recorded data shows that the AMD Ryzen AI 9 HX 375 is the stronger overall processor for most workloads. Its 9 wins in the head-to-head tests include the two Cinebench multi-core tests, integer math, data compression, extended instructions, random string sorting, and the PassMark multithread test. The average benchmark score of 46030 places it in the 89th percentile, and its nearest rivals in the database are the Intel Core Ultra 5 235 (delta of -0.1%), the AMD EPYC 4364P (delta of 0.1%), and the Intel Core i9-13900HX (delta of -0.1%). These rivals are all within a fraction of a percent, indicating the AMD part is balanced at its performance tier.

The Intel Arc G3 EXTREME, with an average score of 36699, sits in the 85th percentile. Its nearest rivals are the Intel Core Ultra 5 225 (delta of -0.6%), AMD Ryzen 5 5600F (delta of -0.7%), AMD Ryzen 5 5500X3D (delta of -0.9%), and AMD Ryzen AI 7 PRO 450 (delta of -1.1%). The Intel part is competitive within its own tier but sits about 20% below the AMD part in average score.

For rendering, multi-core productivity, integer math, and data compression, the AMD Ryzen AI 9 HX 375 is the clear choice from the data. For floating-point math, physics simulation, and single-threaded legacy applications, the Intel Arc G3 EXTREME shows specific advantages. The Intel part also offers higher memory bandwidth and a newer PCIe generation, though with fewer lanes. Users who prioritize thread count, rendering throughput, and integer-heavy workloads should select the AMD part. Users who need floating-point performance, physics simulation, or maximum single-thread PassMark scores should consider the Intel part.

Specification Differences

| Specification | AMD Ryzen AI 9 HX 375 | Intel Arc G3 EXTREME |

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

| Cores | 12 | 14 |

| Threads | 24 | 14 |

| Base Clock | 2.00 GHz | 1.90 GHz |

| Boost Clock | 5.10 GHz | 4.70 GHz |

| TDP | 28 W | 25 W |

| Socket | AMD Socket FP8 | Intel BGA 2540 |

| Architecture | Zen 5 | Not recorded |

| Codename | Strix Point | Panther Lake |

| Generation | Ryzen AI 300 (Zen 5 / Zen 5c) | Arc G3 (Panther Lake) |

| Process Node | 4 nm | 3 nm |

| Foundry | TSMC | Intel |

| Die Size | 233 mm² | Not recorded |

| L1 Cache | 80 KB (per core) | 192 KB (per core) |

| L2 Cache | 1 MB (per core) | 2.5 MB (per core) |

| L3 Cache | 16 MB | 18 MB (shared) |

| Memory Support | DDR5, LPDDR5X | LPDDR5X |

| Memory Bus | Dual-channel | Dual-channel |

| Memory Bandwidth | 89.6 GB/s | 136.5 GB/s |

| PCIe | Gen 4, 16 Lanes (CPU only) | Gen 5, 4 Lanes (CPU only) |

| Integrated Graphics | Radeon 890M | Arc B390 |

| Release Date | 2024-06-30 | 2026-05-27 |

| Part Number | 100-000001682 | SA4R3 |

| Multiplier Unlocked | No | No |

| ECC Memory | No | No |