AMD Ryzen AI 9 365 vs Intel Core Ultra 5 225 Comparison

FAQ

Q: Which processor has more cores and threads?

A: Both processors have 10 cores, but the AMD Ryzen AI 9 365 has 20 threads, while the Intel Core Ultra 5 225 has 10 threads. The AMD part uses simultaneous multithreading; the Intel part does not.

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

A: The largest margin is in PassMark integer math, where the AMD Ryzen AI 9 365 scores 101831 versus 65345 for the Intel Core Ultra 5 225, a 55.8% advantage for AMD.

Q: Which processor wins in Cinebench R23 multi-core?

A: The Intel Core Ultra 5 225 wins Cinebench R23 multi-core with a score of 25891 against 18698 for the AMD Ryzen AI 9 365, a 27.8% lead for Intel.

Q: How do their overall benchmark averages compare?

A: The AMD Ryzen AI 9 365 has a higher average benchmark score of 40048, while the Intel Core Ultra 5 225 averages 36938. The AMD part sits at the 87th percentile of all CPUs, the Intel at the 85th.

Q: What are their nearest rivals according to the database?

A: The AMD Ryzen AI 9 365's closest rival is the AMD Ryzen 7 7700 (delta of -0.1%). The Intel Core Ultra 5 225's closest rival is the AMD Ryzen 5 5600F (delta of 0%).

Q: Which processor has the higher boost clock?

A: The AMD Ryzen AI 9 365 boosts to 5.00 GHz, while the Intel Core Ultra 5 225 boosts to 4.90 GHz. The Intel part has a higher base clock at 3.30 GHz versus 2.00 GHz.

Architecture Differences

The AMD Ryzen AI 9 365 uses the Zen 5 architecture in the Strix Point package, built on a 4 nm process at TSMC. The Intel Core Ultra 5 225 uses the Arrow Lake architecture (Arrow Lake-S) on a 3 nm process, also at TSMC. The AMD die measures 233 mm², while the Intel die is larger at 243 mm². Intel reports a transistor count of 17,800 million; AMD does not disclose a figure in the database.

Cache structures diverge significantly. AMD allocates 80 KB of L1 per core and 1 MB of L2 per core, with a 16 MB L3 cache. Intel allocates 192 KB of L1 per core and 3 MB of L2 per core, with a 20 MB shared L3 cache. The larger per-core L2 and total L3 on the Intel side may explain part of its single-thread dominance.

Memory support also differs. The AMD part supports DDR5 and LPDDR5X over a dual-channel bus with 89.6 GB/s bandwidth. The Intel part supports only DDR5 over a dual-channel bus but reaches 102.4 GB/s. That is a 14.3% higher theoretical memory bandwidth for Intel. PCIe connectivity differs as well: AMD provides Gen 4 with 16 lanes, Intel provides Gen 5 with 20 lanes.

The integrated graphics are different tiers. AMD uses the Radeon 880M, while Intel uses Arc Xe-LPG Graphics with 16 execution units. Both processors are locked (multiplier unlocked: false). AMD targets the mobile segment with Socket FP8 and a 28 W TDP; Intel targets the desktop segment with Socket 1851 and a 65 W TDP. Release dates differ by roughly six months, with AMD releasing at the end of June 2024 and Intel at the start of January 2025.

The Verdict

The data splits this comparison into two distinct profiles. The Intel Core Ultra 5 225 is the stronger single-threaded and multi-core rendering processor. It wins Cinebench R23 multi-core by 27.8% and Cinebench R23 single-core by 45.5%. It also leads in PassMark single-thread by 12.9%. For workloads dominated by Cinebench-style rendering, encryption, floating-point math, or prime number finding, the Intel part is clearly ahead.

The AMD Ryzen AI 9 365 wins the aggregate benchmark average, 40048 versus 36938, and holds a higher percentile rank (87th versus 85th). Its wins are concentrated in integer math (55.8% ahead), data compression (17.1% ahead), and random string sorting (7.8% ahead). The AMD part also delivers 20 threads versus 10, which explains why it stays competitive in multithreaded PassMark (29467 versus 30459, only 3.3% behind) despite losing Cinebench R23 multi-core by a wide margin.

The platform context matters. The AMD chip is a 28 W mobile processor; the Intel chip is a 65 W desktop processor. Despite the desktop part's higher power envelope, the mobile AMD part still wins the overall average benchmark score. For an integrated mobile solution, the AMD Ryzen AI 9 365 provides a broader spread of workload strengths. For a desktop system where single-thread speed and rendering throughput are the priority, the Intel Core Ultra 5 225 is the data-supported choice.

Specification Differences

| Specification | AMD Ryzen AI 9 365 | Intel Core Ultra 5 225 |

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

| Threads | 20 | 10 |

| Base clock | 2.00 GHz | 3.30 GHz |

| Boost clock | 5.00 GHz | 4.90 GHz |

| TDP | 28 W | 65 W |

| Socket | AMD Socket FP8 | Intel Socket 1851 |

| Architecture | Zen 5 | Arrow Lake |

| Codename | Strix Point | Arrow Lake-S |

| Process node | 4 nm | 3 nm |

| Transistors | Not listed | 17,800 million |

| Die size | 233 mm² | 243 mm² |

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

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

| L3 cache | 16 MB | 20 MB (shared) |

| Memory support | DDR5, LPDDR5X | DDR5 |

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

| PCIe | Gen 4, 16 Lanes | Gen 5, 20 Lanes |

| Integrated graphics | Radeon 880M | Arc Xe-LPG Graphics 16EU |

| Market segment | Mobile | Desktop |

| Release date | 2024-06-30 | 2025-01-06 |

| Launch MSRP | Not listed | $246 |

Head-to-Head Benchmarks

The Intel Core Ultra 5 225 wins 11 of the 15 recorded head-to-head comparisons. Its most decisive victories are in PassMark find prime numbers (358 versus 117, a 67.3% lead), Cinebench R23 single-core (3655 versus 1992, a 45.5% lead), and PassMark floating point math (92038 versus 62802, a 31.8% lead). Cinebench R15 single-core also favors Intel, 368 versus 303, a 17.7% margin. PassMark data encryption goes to Intel by 17.9% (22285 versus 18297). PassMark extended instructions, physics, and multithread all go to Intel as well, with margins of 7.5%, 27.2%, and 3.3% respectively.

The AMD Ryzen AI 9 365 wins the remaining four comparisons. The largest is PassMark integer math: 101831 versus 65345, a 55.8% advantage. PassMark data compression goes to AMD by 17.1% (354510 versus 302811). Cinebench R15 multi-core favors AMD by 8.9% (2842 versus 2609). PassMark random string sorting goes to AMD by 7.8% (39447 versus 36590).

The pattern is striking. Intel dominates in floating-point, prime-finding, encryption, and the Cinebench suite, which are heavily single-thread dependent or rely on high per-core throughput. AMD dominates in integer-heavy tasks and compression, where its 20 threads and larger thread count can be fed with parallel work.

Where Each One Wins

The Intel Core Ultra 5 225 is the choice for workloads that reward high single-thread performance and strong floating-point execution. The Cinebench R23 single-core score of 3655 is 83.5% higher than AMD's 1992, and the PassMark floating point score is 46.5% higher. Prime number finding, which is often sensitive to branch prediction and integer latency, shows a 206% lead for Intel. Encryption workloads also prefer Intel, with a 17.9% edge. For desktop users running rendering, scientific simulation, or encryption-heavy tasks, the Intel part consistently delivers higher scores.

The AMD Ryzen AI 9 365 is the better option for integer math and data compression. Its 55.8% lead in PassMark integer math and 17.1% lead in data compression indicate a strong fit for compression tools, archive handling, and integer-heavy code. The 20-thread configuration also narrows the multithread gap to just 3.3% in PassMark, even though the Intel part has a 65 W desktop TDP versus 28 W mobile. The AMD part wins the overall average benchmark score and sits at a higher percentile of all CPUs.

The mobile versus desktop split is also visible in the data. Despite being a 28 W mobile part, the AMD Ryzen AI 9 365 holds a higher average benchmark score than the 65 W desktop Intel part. That suggests the AMD architecture extracts more performance per watt in the measured workloads. The Intel part, however, uses its higher power budget to push single-thread and floating-point performance further. Users on mobile platforms have only one choice here; users building desktops can weigh the Intel part's rendering dominance against AMD's integer and compression strength.