AMD Ryzen AI 9 PRO 465 vs AMD Ryzen AI Embedded P164 Comparison

Head-to-Head Benchmarks

The benchmark data shows a decisive sweep for the AMD Ryzen AI 9 PRO 465, which wins all 11 recorded head-to-head comparisons against the AMD Ryzen AI Embedded P164. The largest margin appears in the prime number search workload, where the PRO 465 scores 126 against 71 for the P164, a 77.5% advantage. That workload is heavily dependent on integer throughput and cache behavior, so the gap reflects more than just clock speed.

In the multithreaded PassMark test, the PRO 465 records 31485 versus 25889, a 21.6% lead. Integer math shows a similar pattern, with 107173 against 87940, a 21.9% delta. Floating-point math lands at 66824 for the PRO 465 versus 55799, a 19.8% gap. These results indicate that the PRO 465's additional two cores and larger L3 cache translate directly into measurable throughput gains across both integer and floating-point instruction streams.

Data compression shows the PRO 465 at 385174 versus 327891, a 17.5% advantage. Encryption workloads favor the PRO 465 even more, with a score of 19308 against 16055, a 20.3% delta. Extended instruction performance is closer, with the PRO 465 leading 26441 to 24193, a 9.3% margin. The physics workload shows a particularly large divergence, 1747 versus 1210, a 44.4% lead for the PRO 465; this suggests that the extra cores and cache hierarchy substantially benefit physics simulation tasks.

Random string sorting, a memory-latency-sensitive test, shows the PRO 465 at 40860 versus 34801, a 17.4% delta. Single-threaded performance is the tightest race: the PRO 465 scores 4168 against 4029, a 3.4% margin. That narrow lead indicates that per-core architecture is very similar, with the PRO 465's advantage coming mostly from its larger pool of cores and shared cache. Across the entire benchmark suite, the PRO 465's average benchmark score is 62498, while the P164 averages 52901, a difference of roughly 18%.

The PRO 465 sits at the 93rd percentile among all CPUs in the database, while the P164 ranks at the 91st percentile. The PRO 465's nearest rivals include the Intel Core Ultra 7 255HX, which scores 62738 and trails by 0.4%, and the AMD Ryzen AI Embedded P185 at 62839, also 0.5% behind. The Intel Core i9-13900KF scores 61841, which is 1.1% lower than the PRO 465. The P164's nearest rivals include the AMD Ryzen 5 9500F at 52873, just 0.1% behind, and the Intel Xeon 634 at 52974, 0.1% ahead. The AMD EPYC 7313P scores 53206, which is 0.6% higher than the P164, and the AMD Ryzen 9 7900X at 53288 is 0.7% higher. These positioning data show that the PRO 465 competes in a higher performance tier, while the P164 sits just below mainstream desktop parts like the Ryzen 9 7900X.

Architecture Differences

Both processors share the same Gorgon Point codename, the same 4 nm process node at TSMC, and the same 233 mm² die size. The PRO 465 belongs to the Ryzen AI PRO 400 generation, while the P164 belongs to the Ryzen AI Embedded generation, but both use a mix of Zen 5 and Zen 5c cores. The PRO 465 provides 10 cores and 20 threads, whereas the P164 provides 8 cores and 16 threads. That two-core difference is the primary architectural distinction and explains most of the multithreaded performance gap.

Both chips run at a 2.00 GHz base clock and a 5.00 GHz boost clock, with a 28 W TDP. Both use AMD Socket FP8 and support DDR5 and LPDDR5X memory over a dual-channel bus with 89.6 GB/s of bandwidth. PCIe support is identical: Gen 4 with 16 lanes from the CPU. Neither processor has an unlocked multiplier.

Cache configurations differ significantly. Each core on both chips has 80 KB of L1 and 1 MB of L2, but the PRO 465 carries 16 MB of L3 cache, while the P164 has only 8 MB. That doubling of shared L3 cache gives the PRO 465 a substantial advantage in workloads that reuse data across cores, such as the compression and physics tests where it leads by 17.5% and 44.4%, respectively.

Integrated graphics also differ. The PRO 465 uses the Radeon 890M, while the P164 uses the Radeon 880M. The P164 supports ECC memory, and the PRO 465 does not. That makes the P164 the better fit for error-sensitive embedded workloads, even though its raw CPU performance is lower. The PRO 465 has a part number of 100-000001862, while the P164's part number is listed as unknown.

The PRO 465 was released on January 4, 2026, and the P164 followed on March 8, 2026. Both are listed as Active in production status and target the Mobile market segment.

FAQ

Q: Which processor has more cores and threads?

A: The AMD Ryzen AI 9 PRO 465 has 10 cores and 20 threads, while the AMD Ryzen AI Embedded P164 has 8 cores and 16 threads.

Q: How much L3 cache does each processor have?

A: The PRO 465 has 16 MB of L3 cache, while the P164 has 8 MB.

Q: Do both processors support ECC memory?

A: No. The P164 supports ECC memory, but the PRO 465 does not.

Q: What is the single-threaded performance difference?

A: The PRO 465 scores 4168 in the PassMark single-thread test, while the P164 scores 4029, a 3.4% lead for the PRO 465.

Q: Which processor has a higher average benchmark score?

A: The PRO 465 averages 62498, while the P164 averages 52901.

Q: Do both processors use the same process node?

A: Yes, both use TSMC's 4 nm process node and have a die size of 233 mm².

Q: What integrated graphics do they use?

A: The PRO 465 uses the Radeon 890M, and the P164 uses the Radeon 880M.

The Verdict

The data clearly favors the AMD Ryzen AI 9 PRO 465 for general-purpose and multithreaded workloads. It wins every benchmark in the comparison, with the largest margins in prime number search, physics, integer math, and multithreaded tests. Its 16 MB L3 cache and 10-core configuration provide a strong foundation for compute-heavy tasks, and its 93rd percentile ranking places it ahead of parts like the Intel Core i9-13900KF and the AMD Ryzen AI Embedded P185.

The AMD Ryzen AI Embedded P164 is not without purpose. Its 8 MB L3 cache and 8-core layout deliver lower absolute performance, but it supports ECC memory, which the PRO 465 lacks. For embedded applications that require error-correcting memory, the P164 is the only viable choice between these two. Its 91st percentile ranking is still respectable, and its average score of 52901 places it close to the AMD Ryzen 5 9500F and Intel Xeon 634.

The choice depends on the workload. For pure compute density, the PRO 465 is the stronger part across every measured metric. For embedded systems where data integrity and ECC support take priority over raw throughput, the P164's feature set justifies its selection. The PRO 465's single-thread lead of 3.4% is modest, so applications that rely primarily on single-core performance will see little difference. Multithreaded and cache-sensitive workloads will see the largest gains from the PRO 465.

Specification Differences

| Specification | AMD Ryzen AI 9 PRO 465 | AMD Ryzen AI Embedded P164 |

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

| Cores | 10 | 8 |

| Threads | 20 | 16 |

| L3 Cache | 16 MB | 8 MB |

| ECC Memory | false | true |

| Integrated Graphics | Radeon 890M | Radeon 880M |

| Generation | Ryzen AI PRO 400 (Zen 5 / Zen 5c) | Ryzen AI Embedded (Zen 5 / Zen 5c) |

| Release Date | 2026-01-04 | 2026-03-08 |

| Part Number | 100-000001862 | unknown |

| Avg Benchmark Score | 62498 | 52901 |

| Percentile vs All CPUs | 93 | 91 |

Shared specifications include a 2.00 GHz base clock, 5.00 GHz boost clock, 28 W TDP, AMD Socket FP8, Gorgon Point codename, 4 nm TSMC process, 233 mm² die size, 80 KB L1 per core, 1 MB L2 per core, DDR5/LPDDR5X memory support, dual-channel memory bus, 89.6 GB/s memory bandwidth, Gen 4 PCIe with 16 lanes, Mobile market segment, and Active production status.