AMD Ryzen AI 7 345 vs Intel Core 5 120UL Comparison

Where Each One Wins

The recorded benchmark data shows a complete sweep for the AMD Ryzen AI 7 345. Across all 15 head-to-head comparisons, the AMD processor holds the higher score, with no benchmark win recorded for the Intel Core 5 120UL. The margin varies widely by workload type, which helps clarify where each part has relative strengths even in a one-sided comparison.

The largest advantages for the AMD part appear in extended instruction throughput and data compression. In the PassMark extended instructions test, the AMD scores 17003 against 5203 for Intel, a delta of 226.8%. That points to heavy SIMD or cryptographic instruction workloads favoring the AMD architecture decisively. Data compression shows a 117.7% lead (237484 versus 109090), indicating strong performance in archiving, database compression, or similar data-dense tasks.

Single-thread performance is also a clear AMD strength. Cinebench R23 single-core shows 1818 versus 1266, a 43.6% advantage. PassMark single-thread confirms this at 86.3% (3875 versus 2080). For applications that rely on per-core responsiveness, such as lightly threaded productivity tools or general desktop interaction, the AMD part delivers substantially higher throughput per thread.

Multi-threaded workloads follow a similar pattern but with a narrower gap in some tests. Cinebench R23 multi-core shows 11461 versus 8974, a 27.7% lead for AMD. However, Cinebench R15 multi-core shows a larger 89.4% delta (1712 versus 904), and PassMark multi-thread shows 88.7% (19927 versus 10558). The variation between Cinebench versions suggests that the AMD part scales better under certain multi-threaded loads, while the Intel part is relatively closer in newer Cinebench versions.

The Intel Core 5 120UL does not win any individual test, but its closest margins are in find prime numbers (31.9% delta) and Cinebench R23 multi-core (27.7%). These remain substantial gaps, yet they are the smallest recorded differences, indicating that the Intel part is least disadvantaged in integer-heavy prime calculation and in a modern multi-core render workload.

Architecture Differences

The two processors come from fundamentally different design points. The AMD Ryzen AI 7 345 uses the Krackan Point codename, part of the Ryzen AI 300 generation built on Zen 5 and Zen 5c cores. It is manufactured on a 4 nm process at TSMC. The Intel Core 5 120UL uses Raptor Lake architecture, specifically the Raptor Lake-PS codename, on a 10 nm process at Intel.

Core counts differ significantly. The AMD part has 6 physical cores and 12 threads, while the Intel part has 10 physical cores and 12 threads. Both present 12 threads to the operating system, but the Intel processor relies on a hybrid core arrangement typical of Raptor Lake, which combines performance and efficiency cores. The AMD processor uses a homogeneous set of Zen 5 and Zen 5c cores, meaning all cores can handle similar instruction workloads but with different clock and power characteristics.

Cache hierarchies also diverge. Both have 80 KB of L1 per core. L2 differs: the AMD part has 1 MB per core, while the Intel part has 1.25 MB per core. L3 cache shows a major difference: the AMD processor has 4 MB total, while the Intel processor has 12 MB shared. This larger L3 on the Intel side may help with repeated data access patterns, but the benchmark results show that the AMD part still leads in every measured test.

Memory support differs in scope. The AMD processor supports DDR5 and LPDDR5X with dual-channel memory and a recorded bandwidth of 89.6 GB/s. The Intel processor supports DDR4 and DDR5 with dual-channel memory, but the database does not list a memory bandwidth figure. The AMD part uses PCIe Gen 4 with 14 lanes (CPU only), while the Intel part uses PCIe Gen 4 with 8 lanes (CPU only). That gives the AMD processor more direct expansion bandwidth for storage or other peripherals.

Integrated graphics also differ. The AMD part uses Radeon 840M, while the Intel part uses Iris Xe Graphics 80EU. Both are capable of display output and basic acceleration, but the benchmark data does not include graphics-specific tests. Market segment classifications differ as well: the AMD part is listed as Mobile, while the Intel part is listed as Desktop. Release dates are close, with the AMD part dated 2025-01-14 and the Intel part dated 2024-04-07.

The Verdict

The data is unambiguous. The AMD Ryzen AI 7 345 outperforms the Intel Core 5 120UL in every single recorded benchmark, with deltas ranging from 27.7% to 226.8%. The AMD part sits at the 81st percentile of all CPUs in the database, while the Intel part sits at the 68th percentile. Average benchmark scores reinforce this: the AMD processor averages 29461, while the Intel processor averages 13594.

For users prioritizing raw compute throughput, the AMD Ryzen AI 7 345 is the clear choice. Its single-thread lead is large enough to affect everyday responsiveness, and its multi-thread lead is substantial across most tested workloads. The extended instructions and compression results suggest particular strength in data processing, scientific computing, or any workload that uses advanced CPU instruction sets.

The Intel Core 5 120UL does have a lower TDP (15 versus 28), which may indicate lower power draw, but the database does not include efficiency measurements. Its larger L3 cache (12 MB versus 4 MB) does not translate into performance wins in the recorded tests. The Intel part also supports DDR4 memory, which could be relevant for systems with existing DDR4 modules, but this is not measured in the benchmark scores.

Given the complete sweep, the AMD Ryzen AI 7 345 is the superior processor in every measured dimension. The Intel Core 5 120UL would only be preferable in a system where its lower TDP, DDR4 compatibility, or desktop socket form factor are decisive requirements, none of which appear in the benchmark performance data.

FAQ

Q: Which processor has more cores?

A: The Intel Core 5 120UL has 10 cores, while the AMD Ryzen AI 7 345 has 6 cores. Both have 12 threads.

Q: In which benchmark does the AMD processor have the largest lead?

A: The largest delta is in PassMark extended instructions, where the AMD Ryzen AI 7 345 scores 17003 versus 5203 for the Intel Core 5 120UL, a 226.8% advantage.

Q: Does the Intel processor win any benchmark?

A: No. The database records 15 head-to-head benchmarks, and the AMD Ryzen AI 7 345 wins all 15.

Q: What is the difference in single-core performance?

A: Cinebench R23 single-core shows 1818 for AMD versus 1266 for Intel, a 43.6% lead. PassMark single-thread shows 3875 versus 2080, an 86.3% lead.

Q: Which processor has more L3 cache?

A: The Intel Core 5 120UL has 12 MB shared L3, while the AMD Ryzen AI 7 345 has 4 MB.

Q: What is the TDP difference?

A: The AMD Ryzen AI 7 345 has a TDP of 28, while the Intel Core 5 120UL has a TDP of 15.

Head-to-Head Benchmarks

The largest single win for the AMD Ryzen AI 7 345 comes in PassMark extended instructions. The AMD score of 17003 dwarfs the Intel score of 5203, a delta of 226.8%. This test typically exercises advanced instruction set extensions, and the AMD Zen 5 architecture shows a dominant advantage here. No other benchmark approaches this margin.

Data compression is the second-largest delta at 117.7%. The AMD part scores 237484, while the Intel part scores 109090. This more than doubles the throughput of the Intel part in compression workloads, which often stress memory bandwidth and integer execution. The AMD part's 89.6 GB/s memory bandwidth likely contributes to this result.

Cinebench R15 single-core shows a 113.4% delta (271 versus 127). This is the oldest Cinebench version in the data, and the AMD part's single-core advantage is particularly pronounced. Cinebench R23 single-core shows a smaller but still large 43.6% delta (1818 versus 1266). The difference between these two Cinebench versions suggests that the AMD part's single-core advantage is consistent across versions, though the magnitude varies.

PassMark multi-thread shows 19927 versus 10558, an 88.7% delta. This aligns closely with Cinebench R15 multi-core's 89.4% delta (1712 versus 904). Both tests reward parallel execution, and the AMD part's 6 Zen 5-based cores outperform the Intel part's 10 Raptor Lake cores in these parallel workloads despite having fewer physical cores.

PassMark single-thread shows 3875 versus 2080, an 86.3% delta. PassMark random string sorting shows 25435 versus 13610, an 86.9% delta. These two tests are nearly identical in margin, indicating that per-thread performance and memory access patterns for sorting both favor the AMD part by similar amounts.

Integer math shows a 66.8% delta (63475 versus 38060), while floating-point math shows a 62% delta (42621 versus 26311). Both are substantial leads for AMD, with integer math being slightly larger. Data encryption shows a 53.7% delta (11814 versus 7685), and physics simulation shows a 34.9% delta (1089 versus 807).

The smallest deltas are in find prime numbers (31.9%, 62 versus 47) and Cinebench R23 multi-core (27.7%, 11461 versus 8974). Even these smallest margins are large enough to be decisive in real-world use. The Intel part's closest relative performance comes in a modern multi-core render workload and in prime number calculation, but it still trails by roughly a quarter to a third.

Specification Differences

The two processors differ across nearly every specification field. The AMD Ryzen AI 7 345 uses 6 cores and 12 threads, while the Intel Core 5 120UL uses 10 cores and 12 threads. Base clocks differ: the AMD part runs at 2.00 GHz, the Intel part at 1.30 GHz. Both boost to 4.60 GHz.

TDP is another clear split. The AMD part is rated at 28, the Intel part at 15. Socket types are incompatible: the AMD part uses AMD Socket FP8, the Intel part uses Intel Socket 1700. Process nodes differ: the AMD part is on 4 nm at TSMC, the Intel part on 10 nm at Intel.

Cache specifications diverge in L2 and L3. Both have 80 KB L1 per core. L2 is 1 MB per core for AMD and 1.25 MB per core for Intel. L3 is 4 MB for AMD and 12 MB shared for Intel. Memory support differs: AMD supports DDR5 and LPDDR5X, Intel supports DDR4 and DDR5. Both use dual-channel memory, but only the AMD part has a recorded memory bandwidth of 89.6 GB/s.

PCIe lane counts differ: the AMD part has Gen 4 with 14 lanes (CPU only), the Intel part has Gen 4 with 8 lanes (CPU only). Integrated graphics are different models: Radeon 840M for AMD, Iris Xe Graphics 80EU for Intel. Market segments differ: AMD is classified as Mobile, Intel as Desktop. Neither processor supports ECC memory, and neither has an unlocked multiplier. The AMD part has a recorded part number of 100-000002122, while the Intel part's part number is listed as unknown.