AMD Ryzen AI 9 465 vs Intel Core Ultra 5 235H Comparison

AMD
AMD

AMD Ryzen AI 9 465

CORE STATE Gorgon Point
CORE SPECS 10 Cores / 20 Threads
CLOCK SPEED 2 Base / 5 GHz Turbo
CACHE 16 MB
MAX TDP 28W
ARCHITECTURE Zen 5
nm
PROCESS 4 nm
LAUNCH DATE 2026
VS
Intel
INTEL

Core Ultra 5 235H

CORE STATE Arrow Lake-H
CORE SPECS 14 Cores / 14 Threads
CLOCK SPEED 2.4 Base / 5 GHz Turbo
CACHE 18 MB (shared)
MAX TDP 28W
ARCHITECTURE Arrow Lake
nm
PROCESS 3 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

cinebench_cinebench_r15_multicore
2,672.5
2,580
cinebench_cinebench_r15_singlecore
247
364
cinebench_cinebench_r23_multicore
17,462.5
25,598
cinebench_cinebench_r23_singlecore
1,996.5
3,613
passmark_data_compression
349,463
301,979
passmark_data_encryption
17,601
23,121
passmark_extended_instructions
24,773
23,354
passmark_find_prime_numbers
124
239
passmark_floating_point_math
62,411
93,509
passmark_integer_math
99,156
74,247
passmark_multithread
28,986
30,091
passmark_physics
1,689
1,985
passmark_random_string_sorting
37,379
36,208
passmark_single_thread
3,750
4,359
passmark_singlethread
3,750
4,359
cinebench_cinebench_r20_multicore
N/A
10,751
cinebench_cinebench_r20_singlecore
N/A
1,517

Analysis: AMD Ryzen AI 9 465 vs Intel Core Ultra 5 235H

Head-to-Head Benchmarks

The recorded data shows a clear split between the AMD Ryzen AI 9 465 and the Intel Core Ultra 5 235H. Across the fifteen head-to-head comparisons, Intel secures ten wins while AMD takes five. The average benchmark score for the AMD part is 43431 against 37522 for Intel, placing the Ryzen AI 9 465 at the 88th percentile of all CPUs and the Core Ultra 5 235H at the 85th percentile. These aggregate figures, however, hide the dramatic divergence in individual workloads.

The largest single victory for AMD comes in PassMark integer math, where the Ryzen AI 9 465 scores 99156 against 74247 for Intel, a 33.5% advantage. This is the most decisive margin in either direction across the entire benchmark set. Data compression also favors AMD heavily: 349463 versus 301979, a 15.7% lead. The AMD processor wins extended instructions by 6.1% (24773 versus 23354) and random string sorting by 3.2% (37379 versus 36208). In Cinebench R15 multi-core, AMD edges ahead with 2672.5 versus 2580, a modest 3.6% margin.

Intel's wins are concentrated in single-threaded performance and floating-point workloads. The most striking result is Cinebench R23 single-core, where the Core Ultra 5 235H scores 3613 against 1996.5, a 44.7% lead. Cinebench R15 single-core shows a 32.1% Intel advantage (364 versus 247), and Cinebench R23 multi-core shows 25598 versus 17462.5, a 31.8% gap. PassMark floating-point math goes to Intel by 33.3% (93509 versus 62411), and prime number finding shows a 48.1% Intel lead (239 versus 124). Data encryption favors Intel by 23.9% (23121 versus 17601), physics by 14.9% (1985 versus 1689), and PassMark multi-thread by 3.7% (30091 versus 28986). PassMark single-thread shows Intel ahead by 14% (4359 versus 3750).

The pattern is consistent: Intel dominates in Cinebench workloads across the board, while AMD excels in integer-heavy PassMark tasks. The multi-core Cinebench results are particularly informative. Despite AMD having 20 threads to Intel's 14, the Intel chip leads by 31.8% in Cinebench R23 multi-core. This suggests the Intel architecture extracts more throughput per thread in rendering-style workloads. The AMD processor's win in Cinebench R15 multi-core is the only Cinebench test it takes, and the margin is small at 3.6%.

PassMark multi-thread is closer than the Cinebench results, with Intel ahead by only 3.7%. This narrower gap indicates that AMD's threading advantage partially compensates for Intel's per-thread strength in this particular test. The data compression result reinforces this interpretation: AMD's 15.7% lead in that workload shows its thread count translating into real throughput gains where parallel integer work dominates.

The Verdict

The benchmark data indicates that the Intel Core Ultra 5 235H is the stronger overall processor for general-purpose computing. Its wins in both Cinebench multi-core and single-core tests, combined with a 14% single-thread PassMark advantage, give it a broader performance profile. The 44.7% lead in Cinebench R23 single-core is the standout figure in the comparison, and the 31.8% multi-core margin in the same test confirms Intel's rendering superiority.

The AMD Ryzen AI 9 465 is the better choice for integer-heavy parallel workloads. Its 33.5% lead in integer math and 15.7% advantage in data compression are substantial, and the 6.1% edge in extended instructions shows strength in SIMD-style integer processing. Users whose workloads involve compression, sorting, or integer computation would see measurable gains from the AMD part. The 3.2% win in random string sorting and the 3.6% Cinebench R15 multi-core result round out its case.

For buyers prioritizing Cinebench-style rendering, encryption, floating-point math, or physics simulations, the Intel part is clearly superior. The 23.9% encryption lead and 33.3% floating-point advantage are decisive. The Intel chip also holds the edge in PassMark multi-thread (3.7%), meaning its overall throughput wins even in a test that should theoretically favor AMD's thread count.

The average benchmark score difference of 5909 points (43431 versus 37522) favors AMD, but this aggregate is skewed by the AMD part's enormous integer math score. The percentile rankings (88 versus 85) also favor AMD, yet the head-to-head win count favors Intel ten to five. The data suggests that the Intel processor is more consistent across varied workloads, while the AMD processor excels in specific integer tasks.

FAQ

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

A: The Intel Core Ultra 5 235H wins ten of the fifteen recorded comparisons, while the AMD Ryzen AI 9 465 wins five.

Q: What is the largest single benchmark margin in either direction?

A: The largest margin is Intel's 48.1% lead in PassMark find prime numbers (239 versus 124). The largest AMD margin is 33.5% in PassMark integer math (99156 versus 74247).

Q: How do the two processors compare in Cinebench R23 multi-core?

A: Intel wins by 31.8%, scoring 25598 against AMD's 17462.5.

Q: Which processor has the higher average benchmark score?

A: The AMD Ryzen AI 9 465 has an average score of 43431, compared to 37522 for the Intel Core Ultra 5 235H.

Q: Do the processors differ in their percentile ranking among all CPUs?

A: Yes, the AMD part sits at the 88th percentile, while the Intel part sits at the 85th percentile.

Q: Which processor is ahead in PassMark single-thread performance?

A: Intel leads by 14%, scoring 4359 versus AMD's 3750.

Specification Differences

The two mobile processors differ in core configuration and thread count. The AMD Ryzen AI 9 465 has 10 cores and 20 threads, while the Intel Core Ultra 5 235H has 14 cores and 14 threads. Intel's chip therefore has more physical cores but no hyperthreading, while AMD's chip has fewer cores but doubles its thread count.

Base clocks differ: AMD starts at 2.00 GHz, Intel at 2.40 GHz. Both boost to 5.00 GHz. Both are locked processors with no unlocked multiplier. They use different sockets: AMD Socket FP8 for the Ryzen AI 9 465, Intel BGA 2049 for the Core Ultra 5 235H. The AMD part has part number 100-000001861, while Intel's is SRQAP.

Memory bandwidth differs substantially. Intel supports 102.4 GB/s against AMD's 89.6 GB/s, though both use dual-channel memory with DDR5 and LPDDR5X support. Neither supports ECC memory. PCIe connectivity also differs: AMD provides Gen 4 with 16 CPU lanes, Intel provides Gen 5 with 8 CPU lanes. The integrated graphics differ as well: AMD uses Radeon 880M, Intel uses Arc Graphics 140T.

The release dates are not aligned. The Intel Core Ultra 5 235H was released on 2025-01-12, while the AMD Ryzen AI 9 465 has a release date of 2025-12-31. Both are active production parts in the mobile market segment.

Architecture Differences

The AMD Ryzen AI 9 465 uses the Zen 5 architecture under the codename Gorgon Point, belonging to the Ryzen AI 400 generation (Zen 5 / Zen 5c). It is built on a 4 nm process at TSMC with a die size of 233 mm². The Intel Core Ultra 5 235H uses the Arrow Lake architecture under the codename Arrow Lake-H, belonging to the Ultra 5 generation (Arrow Lake-H). It is built on a 3 nm process at TSMC, with no die size recorded in the database.

Cache hierarchies differ significantly. AMD provides 80 KB of L1 per core, 1 MB of L2 per core, and 16 MB of L3. Intel provides 192 KB of L1 per core, 3 MB of L2 per core, and 18 MB of shared L3. Intel's larger per-core L1 and L2 caches align with its single-thread performance advantage in the benchmarks. AMD's smaller per-core caches but higher thread count align with its integer throughput wins.

Neither processor uses 3D V-Cache. Both are manufactured by TSMC, though at different nodes. The AMD part's 4 nm process contrasts with Intel's 3 nm process. The die size of 233 mm² is recorded only for AMD; Intel's die size is not available in the database.

Where Each One Wins

The AMD Ryzen AI 9 465 wins in integer-heavy parallel workloads. PassMark integer math shows a 33.5% advantage, making it the strongest AMD result. Data compression follows with a 15.7% lead, and extended instructions add 6.1%. Random string sorting is a narrower win at 3.2%. The Cinebench R15 multi-core result (3.6% lead) is AMD's only multi-core Cinebench victory. These wins suggest workloads involving compression algorithms, integer arithmetic, and data sorting favor the AMD processor.

The Intel Core Ultra 5 235H wins in rendering, floating-point, and single-threaded tasks. Cinebench R23 single-core shows the largest Intel margin at 44.7%, with Cinebench R15 single-core at 32.1% and Cinebench R23 multi-core at 31.8%. PassMark floating-point math gives Intel a 33.3% edge, and find prime numbers shows a 48.1% lead. Data encryption goes to Intel by 23.9%, physics by 14.9%, and PassMark multi-thread by 3.7%. PassMark single-thread shows a 14% Intel advantage.

The Intel processor's broader win distribution makes it the more versatile option. It wins both single-thread and multi-thread Cinebench tests, which are commonly used for rendering and content creation benchmarks. Its wins in encryption and physics extend its utility to security and simulation workloads. The AMD processor's wins are narrower in scope but deeper in margin, particularly in integer math and data compression.

The PassMark multi-thread result (Intel by 3.7%) is notable because it contradicts the Cinebench R23 multi-core result to some degree. While Intel leads by 31.8% in Cinebench R23 multi-core, its PassMark multi-thread lead shrinks to 3.7%. This indicates that AMD's thread advantage helps more in PassMark's multi-thread suite than in Cinebench's rendering workload. The data compression result (AMD by 15.7%) further supports this interpretation: AMD's 20 threads provide tangible benefits in parallel integer tasks.

The average benchmark scores place AMD ahead by 5909 points, and the percentile rankings favor AMD by 3 points. However, the head-to-head win count of ten to five in favor of Intel, combined with the size of Intel's margins in key tests, suggests that Intel's performance is more balanced across the recorded benchmark suite. AMD's high average score is driven largely by its integer math result, which exceeds Intel's by 24909 points. Without that single test, the aggregate picture would shift further toward Intel.

DETAILED SPECIFICATIONS

SPECIFICATION
AI 9 465
Ultra 5 235H
Core Specs
Cores
10
14 +40.0%
Threads
20
14 -30.0%
Base Clock (GHz)
2
2.4 +20.0%
Boost Clock (GHz)
5
5 0.0%
Frequency (GHz)
2
2.4 +20.0%
Turbo Clock (GHz)
5
5 0.0%
Multiplier
20
24 +20.0%
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
18 MB (shared)
Power
TDP (W)
28
28 0.0%
PL1
—
28 W
PL2
—
60 W
Configurable TDP
15-54 W
—
Architecture
Architecture
Zen 5
Arrow Lake
Codename
Gorgon Point
Arrow Lake-H
Generation
Ryzen AI 400 (Zen 5 / Zen 5c)
Ultra 5 (Arrow Lake-H)
Process Size
4 nm
3 nm
Die Size
233 mm²
—
Foundry
TSMC
TSMC
Memory
Memory Support
DDR5, LPDDR5X
DDR5, LPDDR5X
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 BGA 2049
Chipsets
—
WM880, HM870
PCIe
Gen 4, 16 Lanes(CPU only)
Gen 5, 8 Lanes(CPU only)
Intel Hybrid
Hybrid Cores
4 + 6
P-Cores: 4 E-Cores: 10
E-Core Frequency
2000 MHz up to 3.3 GHz
1800 MHz up to 4.4 GHz
LP E-Cores
—
2
AI/NPU
NPU
Yes / 50 TOPS
Yes / 13 TOPS
Graphics
Integrated Graphics
Radeon 880M
Arc Graphics 140T
Other
Market
Mobile
Mobile
Production Status
Active
Active
Part Number
100-000001861
SRQAP
Package
FP8
FC-BGA
Tj Max
100°C
110°C
View Ryzen AI 9 465 Details View Core Ultra 5 235H Details