AMD Ryzen AI 7 450 vs Intel Core Ultra 7 265H Comparison

AMD Ryzen AI 7 450 and Intel Core Ultra 7 265H are both 28 W mobile processors, but the benchmark data shows they are built for different workloads. The AMD part wins only one of the fifteen recorded head-to-head tests, while the Intel part takes the other fourteen. The margin of victory varies wildly, from a 2% gap in one Cinebench test to a 73.4% blowout in another. The choice between them depends entirely on whether the workload favors raw integer throughput or everything else.

Where Each One Wins

The AMD Ryzen AI 7 450 has a single win in the recorded data: PassMark integer math, where it scores 88,531 against the Intel Core Ultra 7 265H's 85,479, a 3.6% advantage. That is the lone bright spot for the AMD chip. Every other benchmark in the database goes to Intel.

The Intel Core Ultra 7 265H dominates across the rest of the suite. Its largest wins come in PassMark find prime numbers (335 vs 89, a 73.4% margin) and PassMark floating point math (109,123 vs 54,447, a 50.1% margin). It also leads by 37.5% in data encryption, 36.8% in physics simulation, and 22.6% in multithreaded throughput. Smaller but still consistent leads appear in single-thread tests, Cinebench multicore, data compression, extended instructions, and random string sorting. The Intel chip is the general-purpose pick; the AMD chip has a narrow integer-math specialty.

Architecture Differences

The two chips take fundamentally different approaches to core layout. The AMD Ryzen AI 7 450 uses 8 cores and 16 threads based on the Zen 5 architecture, with the Gorgon Point codename and the Ryzen AI 400 generation mixing Zen 5 and Zen 5c cores. It is built on a 4 nm process at TSMC with a die size of 195 mm². Cache is split as 80 KB L1 per core, 1 MB L2 per core, and 8 MB total L3.

The Intel Core Ultra 7 265H uses 16 cores and 16 threads based on Arrow Lake, specifically Arrow Lake-H, part of the Core Ultra Series 2. It is built on a 3 nm process at TSMC. Cache is larger across the board: 192 KB L1 per core, 3 MB L2 per core, and 24 MB shared L3. The Intel chip has no die size recorded in the database.

Both support DDR5 and LPDDR5X memory in dual-channel mode, and both support ECC memory. Memory bandwidth differs: the Intel part records 102.4 GB/s against the AMD part's 89.6 GB/s. PCIe support also differs: the AMD chip uses Gen 4 with 16 CPU lanes, while the Intel chip uses Gen 5 with 8 CPU lanes. Integrated graphics are Radeon 860M on the AMD side and Arc Graphics 140T on the Intel side. Both are mobile parts, both are currently active, and neither has an unlocked multiplier.

Head-to-Head Benchmarks

The Cinebench results show a close race in single-core and a clear Intel lead in multi-core. In Cinebench R23 single-core, Intel scores 2,080 against AMD's 2,038, a narrow 2% margin. The R23 multicore test shows Intel at 19,940 against AMD's 18,316, an 8.1% lead. The older Cinebench R15 tests show a larger gap: Intel leads by 9.2% in multicore (2,989 vs 2,713) and by 30% in single-core (307 vs 215). The R15 single-core result is an outlier relative to R23, suggesting the AMD chip's older benchmark result may be penalized by something other than raw clock speed.

PassMark results reveal where the architectures diverge. The Intel chip wins data compression by 5.6% (334,711 vs 315,906), extended instructions by 16.4% (26,805 vs 22,400), random string sorting by 12.5% (40,742 vs 35,648), and single-thread performance by 10% (4,334 vs 3,901). The biggest gaps are in specialized math: floating point nearly doubles (109,123 vs 54,447), and prime number finding is nearly four times faster (335 vs 89). Data encryption shows a 37.5% Intel advantage (26,005 vs 16,247), and physics simulation shows a 36.8% lead (2,497 vs 1,578). Multithread throughput favors Intel by 22.6% (34,027 vs 26,350).

The single AMD win in integer math (88,531 vs 85,479, a 3.6% margin) is the only test where the Ryzen chip's core design produces a better result. The overall average benchmark scores reflect this: Intel sits at 41,621 with an 88th percentile ranking among all CPUs, while AMD sits at 39,485 with an 87th percentile. The Intel chip's nearest rivals include the Core 7 251TE (0.1% behind), the Core i7-14650HX (0.1% ahead), and the Core i7-12850HX (0.4% behind). The AMD chip's nearest rivals include the Ryzen 7 PRO 8840HS (0.3% ahead), the Ryzen 7 PRO 8845HS (0.4% behind), and the Ryzen 7 9800X3D (0.7% ahead). Both chips sit in a tightly packed cluster of similar performers, but the Intel part holds a roughly 5.4% higher average score.

The Verdict

The data points to the Intel Core Ultra 7 265H for almost every workload. It wins fourteen of fifteen recorded tests, including all Cinebench runs, all PassMark math tests except integer math, and all memory-sensitive tests like compression, encryption, and string sorting. The margins are often large: 50.1% in floating point, 73.4% in prime number finding, 37.5% in encryption, and 36.8% in physics. Even where the gap is small, such as the 2% Cinebench R23 single-core margin, Intel still holds the lead.

The AMD Ryzen AI 7 450 is only the better choice for integer-heavy tasks, where its 3.6% lead in PassMark integer math is the sole recorded advantage. That is a narrow niche. For multithreaded productivity, media encoding, encryption, or any workload that stresses floating point or memory throughput, the Intel chip is the stronger part by double-digit margins in most cases. The Intel chip also has a higher average benchmark score (41,621 vs 39,485) and a better percentile ranking (88th vs 87th). The AMD chip's lower core count (8 vs 16) and smaller L3 cache (8 MB vs 24 MB) correlate with its deficits in the multicore and memory-sensitive tests. Unless the target application is specifically integer math, the Intel Core Ultra 7 265H is the pick from the recorded data.

FAQ

Q: Which processor has more cores?

A: The Intel Core Ultra 7 265H has 16 cores, while the AMD Ryzen AI 7 450 has 8 cores. Both have 16 threads.

Q: How large is the performance gap in Cinebench R23 multicore?

A: Intel scores 19,940 against AMD's 18,316, an 8.1% lead for the Intel chip.

Q: In which benchmark does the AMD chip win?

A: The AMD Ryzen AI 7 450 wins PassMark integer math with 88,531 against Intel's 85,479, a 3.6% advantage.

Q: What is the largest single benchmark margin between the two?

A: PassMark find prime numbers shows the biggest gap: Intel scores 335, AMD scores 89, a 73.4% margin for Intel.

Q: Do both processors support the same memory types?

A: Yes, both support DDR5 and LPDDR5X memory. Both also support ECC memory, and both use dual-channel memory buses.

Q: Which chip has higher memory bandwidth?

A: The Intel Core Ultra 7 265H records 102.4 GB/s, while the AMD Ryzen AI 7 450 records 89.6 GB/s.

Specification Differences

| Specification | AMD Ryzen AI 7 450 | Intel Core Ultra 7 265H |

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

| Cores | 8 | 16 |

| Threads | 16 | 16 |

| Base clock | 2.00 GHz | 2.20 GHz |

| Boost clock | 5.10 GHz | 5.30 GHz |

| TDP | 28 W | 28 W |

| Socket | AMD Socket FP8 | Intel BGA 2049 |

| Architecture | Zen 5 | Arrow Lake |

| Codename | Gorgon Point | Arrow Lake-H |

| Generation | Ryzen AI 400 (Zen 5 / Zen 5c) | Ultra 7 (Arrow Lake-H) |

| Process node | 4 nm | 3 nm |

| Foundry | TSMC | TSMC |

| Die size | 195 mm² | Not recorded |

| L1 cache | 80 KB per core | 192 KB per core |

| L2 cache | 1 MB per core | 3 MB per core |

| L3 cache | 8 MB | 24 MB shared |

| Memory bandwidth | 89.6 GB/s | 102.4 GB/s |

| PCIe | Gen 4, 16 lanes (CPU only) | Gen 5, 8 lanes (CPU only) |

| Integrated graphics | Radeon 860M | Arc Graphics 140T |

| Release date | 2026-01-04 | 2025-01-12 |

| Part number | 100-000001868 | SRQAQ |