AMD Ryzen AI 9 HX 370 vs Intel Core Ultra 5 225F Comparison

AMD
AMD

AMD Ryzen AI 9 HX 370

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 225F

CORE STATE Arrow Lake-S
CORE SPECS 10 Cores / 10 Threads
CLOCK SPEED 3.3 Base / 4.9 GHz Turbo
CACHE 20 MB (shared)
MAX TDP 65W
ARCHITECTURE Arrow Lake
nm
PROCESS 3 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

3dmark_16_threads
9,365
N/A
3dmark_2_threads
2,179
N/A
3dmark_4_threads
3,970
N/A
3dmark_8_threads
6,992
N/A
3dmark_max_threads
9,725
N/A
3dmark_single_thread
1,153
N/A
cinebench_cinebench_r15_multicore
3,409.5
2,660
cinebench_cinebench_r15_singlecore
306.5
287
cinebench_cinebench_r23_multicore
21,761
16,467
cinebench_cinebench_r23_singlecore
2,018
1,893
geekbench_multicore
15,242
N/A
geekbench_singlecore
2,407
N/A
passmark_data_compression
445,719
310,843
passmark_data_encryption
22,252
22,648
passmark_extended_instructions
31,653
28,027
passmark_find_prime_numbers
126
352
passmark_floating_point_math
76,892
92,554
passmark_integer_math
122,933
66,417
passmark_multithread
35,148
31,004
passmark_physics
1,881
2,430
passmark_random_string_sorting
48,723
37,325
passmark_single_thread
3,973
4,397
passmark_singlethread
3,973
4,397
cinebench_cinebench_r20_multicore
N/A
11,059
cinebench_cinebench_r20_singlecore
N/A
1,561

Analysis: AMD Ryzen AI 9 HX 370 vs Intel Core Ultra 5 225F

The AMD Ryzen AI 9 HX 370 and Intel Core Ultra 5 225F land remarkably close in overall database standing, yet the recorded data shows they get there in very different ways. The AMD part sits in the 86th percentile of all CPUs with an average benchmark score of 37904, while the Intel part occupies the 85th percentile at 37313. That gap of under one percentile point conceals a head-to-head record that favors AMD nine wins to six, with the margins swinging wildly depending on the workload. One is a mobile chip built for efficiency; the other is a desktop chip built to run at higher power. The benchmarks tell the story of which trade-offs matter.

Head-to-Head Benchmarks

The biggest single result in the entire dataset belongs to AMD: a 85.1 percent win in Passmark integer math, 122933 to 66417. That is not a narrow victory, it is a rout, and it lines up with AMD's 43.4 percent lead in data compression (445719 versus 310843) and its 30.5 percent lead in random string sorting (48723 versus 37325). Any workload dominated by integer throughput and memory-heavy data shuffling belongs firmly to the Ryzen AI 9 HX 370.

The Cinebench suite agrees. In Cinebench R23 multi-core the AMD chip scores 21761 against 16467, a 32.1 percent advantage. The older R15 test shows the same shape: 3409.5 versus 2660, a 28.2 percent gap. Even single-threaded rendering goes to AMD, 2018 versus 1893 in R23 and 306.5 versus 287 in R15, though those margins shrink to single digits at 6.6 and 6.8 percent.

Intel's counterpunch comes in a specific cluster of tests. Passmark single-thread goes to the Core Ultra 5 225F by 9.6 percent, 4397 to 3973, the first sign that raw lightly-threaded responsiveness favors the desktop chip. Passmark physics is Intel's strongest win: 2430 versus 1881, a 22.6 percent margin. Floating point math also belongs to Intel at 92554 versus 76892, a 16.9 percent gap, and prime number finding is nearly a blowout at 352 versus 126, which works out to a 64.2 percent Intel lead. Data encryption is close, with Intel edging ahead 22648 to 22252, a difference of under two percent.

The overall Passmark multithread score lands at 35148 for AMD versus 31004 for Intel, a 13.4 percent AMD advantage. Extended instructions favor AMD by 12.9 percent, 31653 to 28027. The pattern is consistent: AMD dominates sustained multi-core throughput and integer work, Intel takes select scalar and physics-oriented tests plus lightly-threaded responsiveness.

Architecture Differences

These chips come from opposite ends of the market. The Ryzen AI 9 HX 370 is a mobile processor built on AMD's Zen 5 architecture, codenamed Strix Point, fabricated on TSMC's 4 nm process with a 233 mm² die. It belongs to the Ryzen AI 300 generation, which mixes Zen 5 and Zen 5c cores. It carries 12 cores and 24 threads: double the thread count of the Intel part, which explains most of the multi-core results.

The Core Ultra 5 225F is a desktop Arrow Lake-S processor in Intel's Core Ultra Series 2, built on TSMC's 3 nm process with a 243 mm² die and 17,800 million transistors. It has 10 cores and 10 threads, with no hyper-threading equivalent in the data. That thread deficit is the single clearest reason the Cinebench multi-core scores tilt so heavily toward AMD.

Cache philosophies diverge sharply. AMD gives each core 80 KB of L1 and 1 MB of L2, with 16 MB of L3. Intel gives each core 192 KB of L1 and 3 MB of L2, with 20 MB of shared L3. Intel's per-core cache is substantially larger, which may underpin its strong single-thread and physics results.

The platform differences are just as significant. Intel supports PCIe Gen 5 with 20 CPU lanes and reaches 102.4 GB/s of memory bandwidth over dual-channel DDR5. AMD is limited to PCIe Gen 4 with 16 CPU lanes and 89.6 GB/s, though it supports both DDR5 and LPDDR5X. Crucially, the AMD chip includes Radeon 890M integrated graphics while the Intel "F" variant ships with none, requiring a discrete GPU. Their power envelopes sit far apart: a 28 watt TDP for the mobile AMD part against 65 watts for the desktop Intel part. Neither chip has an unlocked multiplier or ECC support, and both are active products from their respective foundry-shared TSMC pipelines.

Where Each One Wins

The Ryzen AI 9 HX 370 is the clear choice for throughput-bound work. Rendering, compression, sorting, integer math, and any application that scales across threads plays to its 24 threads: the 32.1 percent R23 multi-core win and the 85.1 percent integer math win define its territory. Its integrated Radeon 890M graphics also make it self-sufficient in thin systems where a discrete GPU is not an option, and its 28 watt rating means all of this performance arrives at a fraction of the Intel chip's power budget.

The Core Ultra 5 225F wins where lightly-threaded snap and specific numerical kernels matter. Its 9.6 percent single-thread Passmark lead, 22.6 percent physics lead, and 16.9 percent floating point lead show a chip that punches above its thread count in scalar workloads. Its 64.2 percent dominance in prime number finding, while a narrow workload niche, is the largest head-to-head margin anywhere in the dataset. The desktop platform also brings PCIe Gen 5 connectivity and higher memory bandwidth, both relevant when pairing with fast storage and modern discrete graphics.

Specification Differences

The specification sheet diverges on nearly every line. Cores and threads: 12 and 24 for AMD versus 10 and 10 for Intel. Base clock: 2.00 GHz for AMD versus 3.30 GHz for Intel. Boost clock: 5.10 GHz for AMD versus 4.90 GHz for Intel. TDP: 28 watts for AMD versus 65 watts for Intel. Process node: 4 nm for AMD versus 3 nm for Intel. Die size: 233 mm² versus 243 mm². Cache per core: 80 KB L1 and 1 MB L2 on AMD against 192 KB L1 and 3 MB L2 on Intel, with 16 MB of L3 versus 20 MB shared. Memory support spans DDR5 and LPDDR5X on AMD versus DDR5-only on Intel, and bandwidth is 89.6 GB/s versus 102.4 GB/s. PCIe is Gen 4 with 16 lanes versus Gen 5 with 20 lanes. Graphics are integrated on AMD, absent on Intel. The sockets differ entirely: AMD Socket FP8 for a mobile chip, Intel Socket 1851 for a desktop one. Release dates differ too, with the AMD part appearing in mid-2024 and the Intel part in early January 2025. The Intel part's launch MSRP is listed as $231; the AMD part has no recorded launch MSRP in the database.

FAQ

Q: Which CPU is faster overall in the database?

A: The averages are nearly identical: 37904 for the AMD Ryzen AI 9 HX 370 against 37313 for the Intel Core Ultra 5 225F, with percentile placements of 86 versus 85. In direct head-to-head testing, AMD wins 9 of 15 benchmarks.

Q: Which chip is better for multi-threaded workloads like rendering?

A: AMD, decisively. It wins Cinebench R23 multi-core 21761 to 16467 (32.1 percent) and Cinebench R15 multi-core 3409.5 to 2660 (28.2 percent), thanks to 24 threads against Intel's 10.

Q: Does the Intel chip win anything?

A: Yes, six tests. Its best results are prime number finding (352 versus 126), physics (2430 versus 1881), floating point math (92554 versus 76892), and Passmark single-thread (4397 versus 3973).

Q: Which CPU is more power efficient in the recorded data?

A: The AMD chip has a 28 watt TDP versus 65 watts for Intel, while still winning the majority of benchmarks, which indicates substantially better performance per watt for the AMD part.

Q: Do either of these chips include integrated graphics?

A: The AMD Ryzen AI 9 HX 370 includes Radeon 890M graphics. The Intel Core Ultra 5 225F, as an "F" variant, has no integrated graphics and requires a discrete GPU.

Q: Which platform has newer connectivity?

A: Intel. It offers PCIe Gen 5 with 20 CPU lanes and 102.4 GB/s memory bandwidth, versus PCIe Gen 4 with 16 lanes and 89.6 GB/s on AMD.

The Verdict

Pick the Ryzen AI 9 HX 370 for throughput. It wins the head-to-head series nine to six, leads every multi-core rendering test by roughly 28 to 32 percent, and crushes integer and compression workloads by 43 to 85 percent, all within a 28 watt mobile envelope and with Radeon 890M graphics included. Its database neighbors include the AMD Ryzen 7 9700X and Intel Core i9-14901E, both effectively even in average score.

Pick the Core Ultra 5 225F for a desktop build centered on lightly-threaded speed and platform features. It posts the higher Passmark single-thread score, wins physics by 22.6 percent and floating point by 16.9 percent, and sits alongside rivals like the Intel Core i7-13700 and i9-13900HK in overall standing, all while offering PCIe Gen 5, more memory bandwidth, and a bigger per-core cache allocation. Neither chip is a wrong choice; the data simply says the AMD part is the broader performer and the Intel part is the sharper scalar tool on a faster platform.

DETAILED SPECIFICATIONS

SPECIFICATION
AI 9 HX 370
Ultra 5 225F
Core Specs
Cores
12
10 -16.7%
Threads
24
10 -58.3%
Base Clock (GHz)
2
3.3 +65.0%
Boost Clock (GHz)
5.1
4.9 -3.9%
Frequency (GHz)
2
3.3 +65.0%
Turbo Clock (GHz)
5.1
4.9 -3.9%
Multiplier
20
33 +65.0%
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
20 MB (shared)
Power
TDP (W)
28
65 +132.1%
PL1
—
65 W
PL2
—
121 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
No
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: 4
E-Core Frequency
1400 MHz up to 3.3 GHz
2.7 GHz up to 4.4 GHz
P-Core Turbo
—
4.7 GHz
AI/NPU
NPU
Yes / 50 TOPS
—
Graphics
Integrated Graphics
Radeon 890M
—
Other
Market
Mobile
Desktop
Production Status
Active
Active
Launch Price
—
$231
Part Number
100-000000994
SRQD2SRVF9
Package
FP8
FC-LGA18W
Tj Max
100°C
105°C
View Ryzen AI 9 HX 370 Details View Core Ultra 5 225F Details