AMD Ryzen AI 9 HX 375 vs Intel Core Ultra 5 245T Comparison

AMD
AMD

AMD Ryzen AI 9 HX 375

CORE STATE Strix Point
CORE SPECS 12 Cores / 24 Threads
CLOCK SPEED 2 Base / 5.1 GHz Turbo
CACHE 16 MB
MAX TDP 28W
ARCHITECTURE Zen 5
nm
PROCESS 4 nm
LAUNCH DATE 2024
VS
Intel
INTEL

Core Ultra 5 245T

CORE STATE Arrow Lake-S
CORE SPECS 14 Cores / 14 Threads
CLOCK SPEED 2.2 Base / 5.1 GHz Turbo
CACHE 24 MB (shared)
MAX TDP 65W
ARCHITECTURE Arrow Lake
nm
PROCESS 3 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

cinebench_cinebench_r15_multicore
3,334
2,641
cinebench_cinebench_r15_singlecore
301
372
cinebench_cinebench_r23_multicore
21,812
26,208
cinebench_cinebench_r23_singlecore
1,988
3,699
geekbench_multicore
13,957
N/A
geekbench_singlecore
2,084
N/A
passmark_data_compression
404,918
283,812
passmark_data_encryption
20,802
23,656
passmark_extended_instructions
29,269
22,071
passmark_find_prime_numbers
122
324
passmark_floating_point_math
75,153
108,499
passmark_integer_math
121,754
88,676
passmark_multithread
32,916
30,833
passmark_physics
1,819
2,278
passmark_random_string_sorting
44,552
34,931
passmark_single_thread
3,867
4,367
passmark_singlethread
3,867
4,367
cinebench_cinebench_r20_multicore
N/A
11,007
cinebench_cinebench_r20_singlecore
N/A
1,553

Analysis: AMD Ryzen AI 9 HX 375 vs Intel Core Ultra 5 245T

Where Each One Wins

The benchmark split between the AMD Ryzen AI 9 HX 375 and Intel Core Ultra 5 245T is not a simple matter of one chip dominating across the board. The recorded data shows two distinct performance personalities. The AMD processor wins 6 of the 15 head-to-head tests, while the Intel chip takes 9. However, the margin and nature of those wins reveal which workloads favor each design.

The AMD Ryzen AI 9 HX 375 shows its strength in integer-heavy and data-oriented tasks. Its biggest victories come in data compression, where it leads by 42.7 percent, and integer math, where it is ahead by 37.3 percent. Extended instruction workloads also favor the AMD part, with a 32.6 percent advantage. Random string sorting goes to AMD by 27.5 percent. These are tasks that reward high thread counts and efficient parallel execution across many cores. The multithread PassMark score also goes to AMD, though by a smaller 6.8 percent margin, indicating that the AMD part sustains its advantage in mixed parallel workloads even when the gap narrows.

The Intel Core Ultra 5 245T, by contrast, dominates single-threaded and floating-point performance. Its single-core wins are decisive: Cinebench R23 single-core shows a 46.3 percent lead, and Cinebench R15 single-core shows a 19.1 percent lead. PassMark single-thread confirms the pattern with an 11.4 percent advantage. Floating-point math goes to Intel by 30.7 percent, and the prime number finding test shows Intel ahead by 62.3 percent, a massive margin that reflects strong scalar integer throughput per core. Physics simulation also favors Intel by 20.1 percent. Data encryption goes to Intel by 12.1 percent, another workload that benefits from the Intel architecture's per-core efficiency.

The Cinebench multicore results are interesting because they contradict the PassMark multithread result. Intel wins Cinebench R23 multicore by 16.8 percent and Cinebench R15 multicore goes to AMD by 26.2 percent. This split suggests the two chips handle sustained multi-threaded rendering differently, with AMD winning the older R15 test and Intel winning the newer R23 test. The overall pattern is clear: AMD for parallel data processing and integer throughput, Intel for single-threaded responsiveness and floating-point compute.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD Ryzen AI 9 HX 375 uses a hybrid core arrangement based on the Zen 5 architecture, with a mix of Zen 5 and Zen 5c cores in the Strix Point design. It packs 12 cores and 24 threads, meaning every core supports two threads. The base clock is 2.00 GHz with a boost of 5.10 GHz. The chip is built on a 4 nm process at TSMC and occupies a 233 mm² die. Cache is organized as 80 KB of L1 per core, 1 MB of L2 per core, and 16 MB of shared L3. Memory support includes DDR5 and LPDDR5X over a dual-channel bus, yielding 89.6 GB/s of bandwidth. The integrated graphics are Radeon 890M. This is a mobile part on AMD Socket FP8 with a 28 W TDP, and the PCIe interface is Gen 4 with 16 CPU lanes. ECC memory is not supported.

The Intel Core Ultra 5 245T uses the Arrow Lake architecture in the Arrow Lake-S design, part of the Core Ultra Series 2 generation. It has 14 cores but only 14 threads, meaning no hyperthreading on any core. The base clock is 2.20 GHz with the same 5.10 GHz boost as the AMD chip. The process node is 3 nm, also at TSMC, with a die size of 243 mm² and a transistor count of 17,800 million. Cache is significantly larger: 192 KB of L1 per core, 3 MB of L2 per core, and 24 MB of shared L3. Memory support is DDR5 only, still dual-channel, but bandwidth rises to 102.4 GB/s, which is 14.3 percent higher than the AMD part. ECC memory is supported. The integrated graphics are Arc Xe-LPG Graphics with 64 execution units. This is a desktop part on Intel Socket 1851 with a 65 W TDP, and it uses PCIe Gen 5 with 20 CPU lanes. The launch MSRP is $270.

The architecture differences explain much of the benchmark behavior. The AMD chip relies on simultaneous multithreading to extract parallelism from its 12 cores, yielding 24 threads. The Intel chip uses fewer threads but larger caches and a newer process node. The 3 nm process and the massive L2 cache per core (3 MB versus 1 MB) give Intel a strong single-thread advantage. The higher memory bandwidth on Intel also helps floating-point and memory-sensitive workloads. The AMD chip compensates with more threads and a smaller die, which keeps power lower at 28 W versus 65 W.

Head-to-Head Benchmarks

The largest single benchmark margin in the entire comparison belongs to Intel in the prime number test. The Core Ultra 5 245T scores 324 against the Ryzen AI 9 HX 375's 122, a 62.3 percent lead. This test is highly dependent on per-core integer throughput and cache efficiency, and the Intel part's larger L2 cache (3 MB per core) likely explains the result. The second-largest margin is the Cinebench R23 single-core test, where Intel leads by 46.3 percent with a score of 3699 versus 1988. This confirms that Intel's single-core performance is dramatically higher across different benchmark generations.

AMD's largest win is in data compression, where it scores 404918 versus 283812, a 42.7 percent advantage. This test benefits from the AMD chip's 24 threads, which allow it to parallelize compression work across twice as many logical processors as the Intel part's 14 threads. Integer math shows a 37.3 percent lead for AMD (121754 versus 88676), again reflecting the thread-count advantage. Extended instructions go to AMD by 32.6 percent (29269 versus 22071), and random string sorting by 27.5 percent (44552 versus 34931).

The Cinebench multicore results are the most contradictory. In Cinebench R15 multicore, AMD wins with 3334 versus 2641, a 26.2 percent margin. But in Cinebench R23 multicore, Intel wins with 26208 versus 21812, a 16.8 percent margin. The R23 test is longer and more demanding, which may favor the Intel chip's higher sustained performance per core despite fewer threads. The R15 test is shorter and may allow the AMD chip to burst with all 24 threads before thermal limits or power constraints take effect.

The floating-point math test shows Intel ahead by 30.7 percent (108499 versus 75153). This is a significant margin and suggests that the Intel architecture's wider vector units or better FP scheduling give it a clear edge in scientific and engineering workloads. Physics simulation also goes to Intel by 20.1 percent (2278 versus 1819), and data encryption by 12.1 percent (23656 versus 20802). The single-thread PassMark test shows Intel at 4367 versus 3867, an 11.4 percent lead.

The multithread PassMark score is the closest result, with AMD at 32916 versus Intel at 30833, a 6.8 percent margin. This indicates that in a general mixed-thread workload, the AMD chip's thread advantage is partially offset by Intel's per-core strength, resulting in a narrow AMD win. The overall score averages reflect this balance: the AMD chip has an average benchmark score of 46030, placing it at the 89th percentile, while the Intel chip averages 38194, at the 86th percentile.

FAQ

Q: Which processor has more cores and threads?

A: The AMD Ryzen AI 9 HX 375 has 12 cores and 24 threads. The Intel Core Ultra 5 245T has 14 cores but only 14 threads, since it does not support simultaneous multithreading.

Q: What is the single-thread performance difference?

A: The Intel Core Ultra 5 245T leads in all single-thread tests. The largest margin is in Cinebench R23 single-core, where Intel scores 3699 versus AMD's 1988, a 46.3 percent advantage. PassMark single-thread shows Intel at 4367 versus 3867, an 11.4 percent lead.

Q: Which chip wins in multi-threaded rendering?

A: The results depend on the test version. In Cinebench R15 multicore, AMD wins with 3334 versus 2641, a 26.2 percent margin. In Cinebench R23 multicore, Intel wins with 26208 versus 21812, a 16.8 percent margin.

Q: How do the two compare in memory bandwidth?

A: The Intel Core Ultra 5 245T supports 102.4 GB/s of dual-channel DDR5 bandwidth. The AMD Ryzen AI 9 HX 375 supports 89.6 GB/s of dual-channel DDR5 or LPDDR5X bandwidth. Intel's bandwidth is higher by approximately 14.3 percent.

Q: What are the TDP differences?

A: The AMD Ryzen AI 9 HX 375 has a 28 W TDP. The Intel Core Ultra 5 245T has a 65 W TDP. The AMD part is designed for mobile use, while the Intel part is a desktop processor.

Q: Which processor has the higher average benchmark score?

A: The AMD Ryzen AI 9 HX 375 has an average benchmark score of 46030, placing it at the 89th percentile of all CPUs. The Intel Core Ultra 5 245T has an average score of 38194, at the 86th percentile.

The Verdict

The data does not declare a single winner. The AMD Ryzen AI 9 HX 375 is the better choice for workloads that scale with thread count and integer throughput. Its 24 threads deliver clear advantages in data compression, integer math, extended instructions, and random string sorting. The 42.7 percent lead in data compression and 37.3 percent lead in integer math are substantial margins that will matter for database workloads, file archiving, and parallel processing tasks. The PassMark multithread win, even at just 6.8 percent, shows that the AMD part maintains an edge in general mixed-thread scenarios. The 28 W TDP also makes it suitable for mobile platforms where power efficiency is a priority.

The Intel Core Ultra 5 245T is the stronger part for single-threaded and floating-point workloads. The 46.3 percent lead in Cinebench R23 single-core and the 62.3 percent lead in prime number finding are decisive. The 30.7 percent advantage in floating-point math and 20.1 percent lead in physics simulation make it the better fit for scientific computing, numerical analysis, and any application where per-core scalar performance matters more than thread count. The 102.4 GB/s memory bandwidth and 24 MB of shared L3 cache support these strengths. The desktop form factor and 65 W TDP are appropriate for systems where power consumption is less constrained.

The average benchmark scores favor AMD overall, with 46030 versus 38194, a meaningful gap of about 20.5 percent. The percentile ranks confirm this: AMD sits at the 89th percentile versus Intel's 86th. However, the Intel chip's nearest rivals include the AMD Ryzen 7 250 with a delta of -0.1 percent, meaning the two are essentially tied in that comparison group. The AMD chip's nearest rivals include the Intel Core Ultra 5 235 at -0.1 percent and the Intel Core i9-13900HX at -0.1 percent, also essentially tied.

For users whose applications are parallel and integer-heavy, the AMD Ryzen AI 9 HX 375 is the data-supported choice. For users who need maximum single-thread speed and floating-point capability, the Intel Core Ultra 5 245T is clearly ahead. The choice depends entirely on the workload profile, and the benchmark data provides a clear map for making that decision.

DETAILED SPECIFICATIONS

SPECIFICATION
AI 9 HX 375
Ultra 5 245T
Core Specs
Cores
12
14 +16.7%
Threads
24
14 -41.7%
Base Clock (GHz)
2
2.2 +10.0%
Boost Clock (GHz)
5.1
5.1 0.0%
Frequency (GHz)
2
2.2 +10.0%
Turbo Clock (GHz)
5.1
5.1 0.0%
Multiplier
20
22 +10.0%
SMP CPUs
1
1 0.0%
Cache
L1 Cache
80 KB (per core)
192 KB (per core)
L2 Cache
1 MB (per core)
3 MB (per core)
L3 Cache
16 MB
24 MB (shared)
Power
TDP (W)
28
65 +132.1%
PL1
—
35 W
PL2
—
114 W
Configurable TDP
15-54 W
—
Architecture
Architecture
Zen 5
Arrow Lake
Codename
Strix Point
Arrow Lake-S
Generation
Ryzen AI 300 (Zen 5 / Zen 5c)
Ultra 5 (Arrow Lake)
Process Size
4 nm
3 nm
Transistors
—
17,800 million
Die Size
233 mm²
243 mm²
Foundry
TSMC
TSMC
Memory
Memory Support
DDR5, LPDDR5X
DDR5
Memory Bus
Dual-channel
Dual-channel
Memory Bandwidth
89.6 GB/s
102.4 GB/s
ECC Memory
No
Yes
Platform
Socket
AMD Socket FP8
Intel Socket 1851
Chipsets
—
Z890, B860, W880, Q870, H810
PCIe
Gen 4, 16 Lanes(CPU only)
Gen 5, 20 Lanes(CPU only)
Intel Hybrid
Hybrid Cores
4 + 8
P-Cores: 6 E-Cores: 8
E-Core Frequency
2000 MHz up to 3.3 GHz
1700 MHz up to 4.5 GHz
AI/NPU
NPU
Yes / 55 TOPS
—
Graphics
Integrated Graphics
Radeon 890M
Arc Xe-LPG Graphics 64EU
Other
Market
Mobile
Desktop
Production Status
Active
Active
Launch Price
—
$270
Part Number
100-000001682
SRVFF
Package
FP8
FC-LGA18W
Tj Max
100°C
105°C
View Ryzen AI 9 HX 375 Details View Core Ultra 5 245T Details