AMD Ryzen AI 9 365 vs Intel Core Ultra 7 258V Comparison

Head-to-Head Benchmarks

The recorded data presents a decisive overall picture: the AMD Ryzen AI 9 365 wins 14 of the 17 head-to-head benchmark comparisons, while the Intel Core Ultra 7 258V takes 3. The margins, however, tell a more nuanced story than the raw win count.

The most dramatic separation occurs in multi-threaded workloads. In Cinebench R23 multicore, the AMD part scores 18,698 against Intel's 10,301, a 81.5% advantage. The Cinebench R15 multicore test shows a similar gap at 78%, with scores of 2,842 versus 1,596.5. PassMark integer math delivers the largest single delta of the entire comparison: the Ryzen AI 9 365 posts 101,831 versus 42,889, a 137.4% lead. Data compression follows closely, with AMD at 354,510 and Intel at 176,686, a 100.6% difference. Random string sorting also heavily favors AMD at 82.8% (39,447 versus 21,580), and extended instructions show a 70.6% gap (25,113 versus 14,717).

Single-thread results are far closer. The Ryzen AI 9 365 edges ahead in Cinebench R23 single-core by 6.4% (1,992 versus 1,872) and in Geekbench single-core by 7.3% (2,253 versus 2,100). The Cinebench R15 single-core margin narrows to 6.3% (303 versus 285). This pattern suggests that architectural efficiency per clock is comparable, with AMD holding a modest but consistent edge.

The Intel Core Ultra 7 258V claims its wins in specific computational patterns. PassMark find prime numbers shows Intel at 185 versus AMD's 117, a 36.8% advantage. This is the largest Intel victory and indicates a strength in integer-heavy prime calculation loops. PassMark single-thread also goes to Intel, 4,018 versus 3,841, a 4.4% lead. The same score appears under the passmark_singlethread test name, confirming the result. These wins demonstrate that in lightly threaded, instruction-simple workloads, Intel can outperform despite losing the broader single-core comparisons.

Intermediate workloads show AMD ahead but with smaller margins. Geekbench multicore favors AMD by 47.6% (13,760 versus 9,325). PassMark multithread shows a 56% lead (29,467 versus 18,887). Floating point math is closer at 9.5% (62,802 versus 57,372), and physics simulation shows only an 8.9% edge (1,704 versus 1,565). Data encryption completes the AMD win list at 35.2% (18,297 versus 13,534).

The average benchmark scores place these processors in different competitive strata. The Ryzen AI 9 365 averages 40,048 and sits at the 87th percentile of all CPUs, with its nearest rival being the AMD Ryzen 7 7700 at a 0.1% difference. The Core Ultra 7 258V averages 20,454 at the 74th percentile, with the AMD Ryzen 5 5600 as its closest comparable at a 0.1% gap. This nearly 2x difference in average score underscores the multi-thread dominance of the AMD part.

Architecture Differences

The two processors take fundamentally different design approaches. The AMD Ryzen AI 9 365 uses the Zen 5 architecture under the Strix Point codename, part of the Ryzen AI 300 generation that combines Zen 5 and Zen 5c cores. It is built on a 4 nm process at TSMC with a die size of 233 mm². The Intel Core Ultra 7 258V uses the Lunar Lake architecture, also fabricated by TSMC but on a 3 nm node. Intel's die size is not recorded in the database.

Core and thread counts diverge sharply. AMD provides 10 cores and 20 threads, while Intel provides 8 cores and 8 threads. The absence of Hyper-Threading on Lunar Lake means Intel's thread count equals its core count, a notable departure from typical Intel designs. Base clocks are close: 2.00 GHz for AMD and 2.20 GHz for Intel. Boost clocks favor AMD at 5.00 GHz versus 4.80 GHz.

Cache hierarchies reflect different philosophies. AMD allocates 80 KB of L1 per core and 1 MB of L2 per core, with 16 MB of shared L3. Intel allocates a larger 192 KB of L1 per core and 2.5 MB of L2 per core, but only 12 MB of shared L3. The larger per-core caches on Intel may explain its prime number and single-thread wins, while AMD's larger L3 pool and additional cores drive its multi-thread advantage.

Memory support differs in type and bandwidth. AMD supports both DDR5 and LPDDR5X over a dual-channel bus, delivering 89.6 GB/s. Intel supports only LPDDR5X, also dual-channel, but delivers 136.5 GB/s, a 52% higher memory bandwidth. This bandwidth advantage could matter in memory-bound tasks, though the recorded benchmarks do not isolate that variable.

PCIe connectivity presents a clear trade-off. AMD offers PCIe Gen 4 with 16 CPU lanes, while Intel offers PCIe Gen 5 with only 4 CPU lanes. The AMD configuration supports more simultaneous devices, while Intel's fewer lanes run at a faster generation. Both processors use non-ECC memory, have locked multipliers, and target the mobile segment with active production status.

Integrated graphics differ as well. AMD pairs its CPU with the Radeon 880M, while Intel uses the Arc 140V. The database records no direct graphics benchmarks for either, so no performance conclusion is drawn from the data. Power targets are recorded: AMD has a 28 W TDP, while Intel operates at 17 W. This 11 W difference aligns with Intel's lower core count and suggests different thermal envelopes for system designers.

Release timing shows AMD shipping earlier, with a release date of 2024-06-30, followed by Intel on 2024-09-23. Both are current, active products. AMD uses the FP8 socket, Intel uses BGA 2833. Neither has a recorded launch MSRP in the database.

FAQ

Q: Which processor wins more head-to-head benchmark comparisons?

A: The AMD Ryzen AI 9 365 wins 14 of the 17 recorded head-to-head tests, while the Intel Core Ultra 7 258V wins 3.

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

A: PassMark integer math shows the biggest gap, with the AMD part scoring 101,831 versus Intel's 42,889, a 137.4% difference in AMD's favor.

Q: Does Intel win any benchmarks?

A: Yes. The Intel Core Ultra 7 258V wins PassMark find prime numbers by 36.8% (185 versus 117) and PassMark single-thread by 4.4% (4,018 versus 3,841).

Q: How do the average benchmark scores compare?

A: The AMD Ryzen AI 9 365 averages 40,048 and ranks at the 87th percentile. The Intel Core Ultra 7 258V averages 20,454 and ranks at the 74th percentile.

Q: What memory bandwidth does each processor support?

A: The Intel part supports 136.5 GB/s with LPDDR5X, while the AMD part supports 89.6 GB/s with DDR5 and LPDDR5X.

Q: How do core counts and threads differ?

A: The AMD Ryzen AI 9 365 has 10 cores and 20 threads. The Intel Core Ultra 7 258V has 8 cores and 8 threads.

The Verdict

The benchmark data supports a clear split by workload type. For multi-threaded, heavily parallel tasks such as video rendering, data compression, encryption, and integer math, the AMD Ryzen AI 9 365 is the stronger choice. Its 20 threads and 16 MB of L3 cache translate directly into leads ranging from 35.2% in encryption to 137.4% in integer math. The Cinebench R23 multicore result of 18,698 versus 10,301 confirms that the AMD part sustains high throughput where core scaling matters.

For lightly threaded workloads, the choice is less obvious. Intel's PassMark single-thread win at 4,018 versus 3,841 shows that in simple single-threaded loops, Intel can pull ahead. The prime number test result, where Intel leads by 36.8%, further reinforces this niche. However, AMD still wins the recorded single-core benchmarks in Cinebench and Geekbench, meaning Intel's single-thread advantage is limited to specific instruction patterns rather than general single-core performance.

The 87th versus 74th percentile ranking places the AMD part in a higher overall performance class. Its nearest rivals include the AMD Ryzen 7 7700 and Intel Core i9-13905H, both desktop-class or high-end mobile parts. The Intel part's nearest rivals are the AMD Ryzen 5 5600 and Ryzen 5 8500G, indicating it competes at a lower performance tier. The data does not record any price information, so no cost-based judgment is made here.

The 17 W TDP of the Intel part versus 28 W for AMD suggests Intel may fit into thinner, lower-power designs. The database does not record thermal or battery measurements, so the practical impact of this difference is not quantified. System designers weighing power budgets would look at this gap, but benchmark data alone cannot determine the winner for such use cases.

Specification Differences

| Specification | AMD Ryzen AI 9 365 | Intel Core Ultra 7 258V |

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

| Cores | 10 | 8 |

| Threads | 20 | 8 |

| Base clock | 2.00 GHz | 2.20 GHz |

| Boost clock | 5.00 GHz | 4.80 GHz |

| TDP | 28 W | 17 W |

| Socket | AMD Socket FP8 | Intel BGA 2833 |

| Architecture | Zen 5 | Lunar Lake |

| Codename | Strix Point | Lunar Lake |

| Generation | Ryzen AI 300 (Zen 5 / Zen 5c) | Ultra 7 (Lunar Lake) |

| Process node | 4 nm | 3 nm |

| Foundry | TSMC | TSMC |

| 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 | 12 MB shared |

| Memory support | DDR5, LPDDR5X | LPDDR5X |

| Memory bandwidth | 89.6 GB/s | 136.5 GB/s |

| PCIe | Gen 4, 16 lanes | Gen 5, 4 lanes |

| Integrated graphics | Radeon 880M | Arc 140V |

| Release date | 2024-06-30 | 2024-09-23 |

| Part number | 100-000001530 | SRPMNSRPMT |

Where Each One Wins

The AMD Ryzen AI 9 365 wins across the majority of recorded workloads. Its 100.6% lead in data compression and 82.8% lead in random string sorting point to strengths in data processing tasks. The 137.4% integer math margin and 70.6% extended instructions margin indicate heavy number-crunching workloads favor AMD. Video encoding, 3D rendering, and software compilation, represented by the Cinebench and Geekbench multicore results, all favor AMD by margins from 47.6% to 81.5%. The physics simulation win at 8.9% and floating point win at 9.5% show AMD also holds a moderate edge in scientific and simulation workloads. The 56% multithread lead and 35.2% encryption lead round out a comprehensive multi-thread advantage.

The Intel Core Ultra 7 258V wins in three specific areas. The PassMark find prime numbers test, where it leads 185 to 117, indicates a strength in tight integer loops that fit within its larger per-core L2 cache. The PassMark single-thread win at 4,018 versus 3,841 shows that for a narrow set of single-threaded operations, Intel's architecture delivers higher throughput. These wins are narrow in scope: the prime number test is a single specialized workload, and the single-thread score is one metric among several. The data does not show any multi-thread workload where Intel leads.

The broader database context supports this split. The AMD part's average score of 40,048 places it near the AMD Ryzen 7 7700, a desktop processor, while the Intel part's 20,454 average aligns with the AMD Ryzen 5 5600. This suggests that for any workload that scales across cores, the AMD Ryzen AI 9 365 is the clear choice from the recorded data. For workloads that are inherently single-threaded and simple, the Intel Core Ultra 7 258V offers a specific but limited advantage.