AMD Ryzen AI 9 HX 375 vs Intel Core Ultra 5 235 Comparison

Intel Core Ultra 5 235 and AMD Ryzen AI 9 HX 375 are two very different processors that end up in the same performance bracket. The data shows their average benchmark scores are nearly identical—46062 for the Intel part versus 46030 for the AMD part, a delta of just 0.1%. Both sit at the 89th percentile of all CPUs. Yet the way they achieve that parity could not be more different, and that distinction matters greatly depending on what you are doing. The Intel chip is a desktop part with fewer threads and higher power limits; the AMD chip is a mobile part with far more threads and a much lower power envelope. Below is a breakdown of where each processor wins, based strictly on the benchmark data.

Where Each One Wins

The Intel Core Ultra 5 235 wins 10 of the 15 head-to-head benchmark comparisons, while the AMD Ryzen AI 9 HX 375 wins 5. The split is not random—it follows a clear pattern. Intel dominates in single-threaded performance, encryption, physics calculations, and floating-point math. The AMD chip’s wins come in multi-threaded Cinebench workloads, integer math, and data compression.

For general desktop responsiveness, the Intel chip is the clear winner. Its PassMark single-thread score of 4516 is 16.8% higher than the AMD chip’s 3867. That advantage carries over into Cinebench R23 single-core, where Intel scores 2085 versus AMD’s 1988, a 4.9% lead. If your work involves a mix of light tasks, spreadsheet churn, or anything that relies on a single fast core, the Intel processor will feel snappier.

For heavy multi-threaded rendering, the AMD chip takes the crown. Its Cinebench R23 multi-core score of 21812 crushes Intel’s 14769, a 32.3% deficit for Intel. The same story plays out in Cinebench R15 multi-core, where AMD scores 3334 versus Intel’s 1488—a massive 55.4% difference. The AMD chip’s 24 threads versus Intel’s 14 threads is the primary driver here. If you are rendering video or compiling code, the AMD processor is the one to pick.

The Intel chip also wins in specialized math workloads. Its PassMark floating-point math score of 117951 is 56.9% higher than AMD’s 75153, and its prime number finding score of 371 is 204.1% higher than AMD’s 122. Encryption is another Intel stronghold: the PassMark data encryption score of 29293 beats AMD’s 20802 by 40.8%. The AMD chip counters in integer math, scoring 121754 versus Intel’s 87948, a 27.8% advantage.

FAQ

Q: Which processor has the higher average benchmark score?

A: The Intel Core Ultra 5 235 has an average benchmark score of 46062, while the AMD Ryzen AI 9 HX 375 scores 46030. The delta is just 0.1%, making them statistically tied.

Q: Is the AMD Ryzen AI 9 HX 375 better for multi-threaded workloads?

A: Yes, decisively. It wins Cinebench R15 multi-core by 55.4% (3334 vs 1488) and Cinebench R23 multi-core by 32.3% (21812 vs 14769). This aligns with its 24 threads versus Intel’s 14.

Q: Does the Intel Core Ultra 5 235 have any single-thread advantage?

A: Yes. It wins Cinebench R23 single-core by 4.9% (2085 vs 1988) and PassMark single-thread by 16.8% (4516 vs 3867).

Q: Which chip is better for data encryption tasks?

A: The Intel Core Ultra 5 235 is far ahead. Its PassMark data encryption score is 29293, which is 40.8% higher than the AMD chip’s 20802.

Q: How do they compare in PassMark multithread performance?

A: Intel wins this one, scoring 37816 versus AMD’s 32916, a 14.9% advantage. This is notable because AMD wins most other multi-thread tests.

Q: Are these chips close enough to be interchangeable?

A: In average score, yes—the 0.1% delta is negligible. But the distribution of wins is lopsided. Intel wins 10 benchmarks; AMD wins 5. They are not interchangeable for specific tasks.

Head-to-Head Benchmarks

The single biggest win for the AMD Ryzen AI 9 HX 375 is in Cinebench R15 multi-core. AMD scores 3334 versus Intel’s 1488, a delta of -55.4% from Intel’s perspective. That is a staggering margin, more than double the performance. Cinebench R23 multi-core is also a decisive AMD win, 21812 versus 14769, a 32.3% gap.

The Intel Core Ultra 5 235 has its own blowout victories. In PassMark’s find prime numbers test, Intel scores 371 against AMD’s 122, a 204.1% advantage. That is the largest single-test margin in either direction. Intel also wins floating-point math by 56.9% (117951 vs 75153) and physics by 41.3% (2570 vs 1819).

The pattern flips entirely when moving to integer-heavy workloads. AMD wins PassMark integer math by 27.8%, scoring 121754 versus Intel’s 87948. AMD also wins data compression by 3.5% (404918 vs 390711) and Cinebench R15 single-core by 30.2% (301 vs 210). However, in the more modern Cinebench R23 single-core test, Intel wins by 4.9% (2085 vs 1988).

Intel sweeps most of the remaining PassMark tests. It wins data encryption by 40.8%, extended instructions by 11.9%, multithread by 14.9%, random string sorting by 9.9%, and single-thread by 16.8%. The final tally is 10 wins for Intel and 5 for AMD, but the AMD wins are concentrated in the most heavily multi-threaded tests.

Specification Differences

The core counts differ significantly. The Intel Core Ultra 5 235 has 14 cores and 14 threads, meaning no hyperthreading. The AMD Ryzen AI 9 HX 375 has 12 cores but 24 threads, thanks to simultaneous multithreading. That explains the multi-core divergence.

Clock speeds tell a story of their own. Intel’s base clock is 3.40 GHz with a boost of 5.00 GHz. AMD’s base clock is much lower at 2.00 GHz, but its boost reaches 5.10 GHz. The Intel chip sustains higher clocks at baseline, while the AMD chip relies on boosting.

Thermal design power (TDP) is a major differentiator. The Intel part is rated at 65W, while the AMD part is rated at just 28W. That is a 37W difference, making the AMD chip far more efficient on paper for mobile use. The Intel chip is a desktop part on Socket 1851; the AMD chip is a mobile part on Socket FP8.

Memory support differs as well. Intel supports DDR5 with a dual-channel bus and a memory bandwidth of 102.4 GB/s. AMD supports both DDR5 and LPDDR5X, also dual-channel, but with a lower bandwidth of 89.6 GB/s. PCIe generations differ: Intel uses Gen 5 with 20 lanes, while AMD uses Gen 4 with 16 lanes.

Architecture Differences

The Intel Core Ultra 5 235 is built on Arrow Lake architecture, part of the Core Ultra Series 2, and manufactured on a 3nm process by TSMC. It has 17,800 million transistors on a 243 mm² die. The cache layout includes 192 KB L1 per core, 3 MB L2 per core, and 24 MB of shared L3 cache. Its integrated graphics are Arc Xe-LPG with 24 execution units. The market segment is desktop.

The AMD Ryzen AI 9 HX 375 uses Zen 5 architecture, codenamed Strix Point, and is part of the Ryzen AI 300 generation with a mix of Zen 5 and Zen 5c cores. It is built on a 4nm process, also by TSMC, with a die size of 233 mm²—slightly smaller than Intel’s. The cache is more modest: 80 KB L1 per core, 1 MB L2 per core, and 16 MB of L3. Its integrated graphics are the Radeon 890M, and it targets the mobile segment. The transistor count is not listed for the AMD chip.

The process node difference (3nm for Intel vs 4nm for AMD) is notable, but the architecture goals are different. Intel’s Arrow Lake emphasizes high clocks and single-core performance. AMD’s Strix Point emphasizes thread count and efficiency, with a TDP less than half of Intel’s. The Intel chip also has a larger L3 cache (24 MB vs 16 MB), which may contribute to its wins in memory-sensitive tasks like encryption and string sorting.

The Verdict

If you are building a desktop and want the fastest single-threaded response, the Intel Core Ultra 5 235 is the pick. It wins 10 of 15 benchmarks, including decisive margins in encryption (40.8%), floating-point math (56.9%), and physics (41.3%). Its PassMark multithread score of 37816 also beats AMD’s 32916, despite having 10 fewer threads. The higher TDP of 65W is a non-issue on desktop, and the Gen 5 PCIe support gives future expansion headroom.

If you are buying a laptop or prioritizing multi-threaded rendering, the AMD Ryzen AI 9 HX 375 is the answer. Its 24 threads deliver a 32.3% win in Cinebench R23 multi-core and a 55.4% win in Cinebench R15 multi-core. The 28W TDP makes it suitable for thin-and-light machines, and the lower memory bandwidth of 89.6 GB/s is less relevant for mobile workloads. The AMD chip also wins integer math by 27.8%, which matters for code compilation and data processing.

The average scores are a tie, but the workloads are not. Intel wins more tests, but AMD wins the tests that scale with core count. Choose based on your primary use case, not the aggregate. The data does not support a universal winner.