AMD Ryzen AI 9 365 vs AMD Ryzen AI Max PRO 390 Comparison

AMD
AMD

AMD Ryzen AI 9 365

CORE STATE Strix 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 2024
VS
AMD
AMD

Ryzen AI Max PRO 390

CORE STATE Strix Halo
CORE SPECS 12 Cores / 24 Threads
CLOCK SPEED 3.2 Base / 5 GHz Turbo
CACHE 64 MB (shared)
MAX TDP 55W
ARCHITECTURE Zen 5
nm
PROCESS 4 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

cinebench_cinebench_r15_multicore
2,842
3,918
cinebench_cinebench_r15_singlecore
303
303
cinebench_cinebench_r23_multicore
18,698
24,828
cinebench_cinebench_r23_singlecore
1,992
1,976
geekbench_multicore
13,760
N/A
geekbench_singlecore
2,253
N/A
passmark_data_compression
354,510
506,170
passmark_data_encryption
18,297
26,027
passmark_extended_instructions
25,113
40,888
passmark_find_prime_numbers
117
323
passmark_floating_point_math
62,802
94,666
passmark_integer_math
101,831
148,508
passmark_multithread
29,467
42,912
passmark_physics
1,704
2,773
passmark_random_string_sorting
39,447
55,248
passmark_single_thread
3,841
3,967
passmark_singlethread
3,841
3,967

Analysis: AMD Ryzen AI 9 365 vs AMD Ryzen AI Max PRO 390

Head-to-Head Benchmarks

The data shows a decisive performance advantage for the AMD Ryzen AI Max PRO 390 across the benchmark suite, winning 13 of the 15 head-to-head comparisons. The Ryzen AI 9 365 takes only two wins, both in single-core tests, and by narrow margins. The largest single gap is in PassMark's find prime numbers test, where the Max PRO 390 scores 323 against the 365's 117, a 63.8% deficit for the smaller chip. That kind of separation indicates a major throughput difference in integer-heavy workloads.

Cinebench results reinforce the trend. In Cinebench R23 multicore, the Max PRO 390 scores 24,828 versus 18,698 for the 365, a 24.7% lead. The R15 multicore test shows a similar story: 3,918 for the Max PRO 390 versus 2,842 for the 365, a 27.5% advantage. These are not marginal wins; they reflect a substantial multicore performance gap between the two parts.

Single-core results are nearly tied. In Cinebench R15 single-core, both chips record exactly 303. In R23 single-core, the 365 edges ahead with 1,992 against 1,976, a 0.8% lead. The PassMark single-thread test gives the Max PRO 390 a slight edge, 3,967 versus 3,841, a 3.2% difference. The database records the R15 single-core tie as a win for the 365 with a delta of 0, so the single-core story is essentially a wash, with the 365 holding a razor-thin advantage in one test and the Max PRO 390 holding a modest one in another.

PassMark's math and encryption workloads show the Max PRO 390 pulling away decisively. In floating-point math, it scores 94,666 against 62,802, a 33.7% lead. Integer math shows 148,508 versus 101,831, a 31.4% gap. Extended instructions (which include AVX-style workloads) favor the Max PRO 390 by 38.6%, with scores of 40,888 and 25,113. Data encryption also favors the larger chip, 26,027 versus 18,297, a 29.7% difference.

Memory and compression tasks follow the same pattern. PassMark data compression scores 506,170 for the Max PRO 390 and 354,510 for the 365, a 30% gap. Random string sorting shows 55,248 versus 39,447, a 28.6% difference. The multithread PassMark score is 42,912 for the Max PRO 390 versus 29,467 for the 365, a 31.3% gap. Physics simulation in PassMark favors the Max PRO 390 by 38.6%, with 2,773 against 1,704.

The average benchmark score puts the Max PRO 390 at 63,765, which places it in the 93rd percentile of all CPUs in the database. The 365 averages 40,048, sitting in the 87th percentile. The nearest rival for the Max PRO 390 is the Intel Core i9-13900KS, which scores 64,051, a 0.4% difference. For the 365, the nearest rival is the AMD Ryzen 7 7700 at 40,081, a 0.1% gap. These percentile and rival positions confirm that the Max PRO 390 is competing in a higher performance tier entirely.

Architecture Differences

Both processors use AMD's Zen 5 architecture and are built on TSMC's 4 nm process node, but they diverge sharply in configuration. The Ryzen AI 9 365 is a Strix Point part from the Ryzen AI 300 generation, combining Zen 5 and Zen 5c cores. It has 10 cores and 20 threads. The Ryzen AI Max PRO 390 is a Strix Halo part from the Ryzen AI Max PRO generation, using only Zen 5 cores. It has 12 cores and 24 threads.

The cache hierarchy differs meaningfully. Both chips provide 80 KB of L1 per core and 1 MB of L2 per core, but the L3 cache is where the gap appears. The 365 has 16 MB of L3, while the Max PRO 390 has 64 MB shared L3. That fourfold difference in L3 capacity is a major factor behind the multicore and math workload results, since larger shared caches reduce memory latency for frequently accessed data.

Memory support also separates the two. The 365 supports DDR5 and LPDDR5X over a dual-channel memory bus, providing 89.6 GB/s of bandwidth. The Max PRO 390 supports only LPDDR5X but uses a quad-channel memory bus, delivering 256.0 GB/s. That is nearly three times the bandwidth, which directly explains the compression, encryption, and sorting advantages in the benchmark data. The Max PRO 390 also supports ECC memory, while the 365 does not.

The integrated graphics differ as well. The 365 uses the Radeon 880M, while the Max PRO 390 uses the Radeon 8050S. Both are integrated solutions, but the Max PRO 390's higher memory bandwidth and larger cache pool likely benefit GPU-bound tasks. The database does not include separate GPU benchmark scores, so the comparison here is limited to CPU-side results.

Sockets and power envelopes are distinct. The 365 uses AMD Socket FP8 with a 28 W TDP. The Max PRO 390 uses AMD Socket FP11 with a 55 W TDP. That higher power envelope, combined with more cores and more cache, accounts for the consistent performance lead. The boost clocks are identical at 5.00 GHz, but the base clock differs: the 365 runs at 2.00 GHz, while the Max PRO 390 runs at 3.20 GHz.

Both parts use PCIe Gen 4 with 16 lanes for the CPU. Neither has an unlocked multiplier. The 365 has a die size of 233 mm², while the Max PRO 390's die size is not recorded in the database. The 365 launched on 2024-06-30, while the Max PRO 390 launched on 2025-01-05, roughly six months later.

Where Each One Wins

The Ryzen AI 9 365 wins only in single-core tests, and even there the margins are small. In Cinebench R23 single-core, it leads by 0.8%. The R15 single-core test is a tie at 303, recorded as a win for the 365 due to the delta of 0. For workloads that depend on single-thread responsiveness, light scripting, or legacy applications that cannot use many cores, the 365 is functionally equivalent to the Max PRO 390, with a negligible edge in one test.

The Ryzen AI Max PRO 390 wins everywhere else. The largest margins come in prime number finding (63.8% ahead), extended instructions and physics (both 38.6% ahead), floating-point math (33.7%), integer math (31.4%), multithread (31.3%), data compression (30%), data encryption (29.7%), and random string sorting (28.6%). In Cinebench multicore tests, it leads by 24.7% (R23) and 27.5% (R15). These results point to clear advantages in rendering, scientific computation, encryption, compression, and any workload that scales across cores or benefits from high memory bandwidth.

The single-thread PassMark test gives the Max PRO 390 a 3.2% edge, which is the only single-core test where it holds the lead. This suggests that even in lightly threaded scenarios, the Max PRO 390's higher base clock (3.20 GHz versus 2.00 GHz) provides a slight real-world advantage, despite the R23 result favoring the 365.

For users running long, multi-threaded jobs such as video encoding, 3D rendering, or batch data processing, the Max PRO 390 is the clear choice. For users whose primary workloads are single-threaded and latency-sensitive, the 365 holds its own, but the database does not show any scenario where it meaningfully outpaces the larger chip.

FAQ

Q: Which processor has more cores and threads?

A: The AMD Ryzen AI Max PRO 390 has 12 cores and 24 threads, while the AMD Ryzen AI 9 365 has 10 cores and 20 threads.

Q: How do the two compare in single-core performance?

A: They are nearly identical. In Cinebench R15 single-core, both score exactly 303. In R23 single-core, the 365 leads with 1,992 versus 1,976, a 0.8% edge. In PassMark single-thread, the Max PRO 390 leads with 3,967 versus 3,841, a 3.2% edge.

Q: What is the biggest benchmark gap between the two?

A: The largest gap is in PassMark's find prime numbers test, where the Max PRO 390 scores 323 and the 365 scores 117, a 63.8% difference in favor of the Max PRO 390.

Q: Do the two chips support the same memory types and bandwidth?

A: No. The 365 supports DDR5 and LPDDR5X over a dual-channel bus with 89.6 GB/s bandwidth. The Max PRO 390 supports only LPDDR5X over a quad-channel bus with 256.0 GB/s bandwidth. The Max PRO 390 also supports ECC memory, while the 365 does not.

Q: How do their cache sizes differ?

A: Both have 80 KB L1 and 1 MB L2 per core, but the L3 cache differs: the 365 has 16 MB, while the Max PRO 390 has 64 MB shared.

Q: What are their power envelopes?

A: The 365 has a 28 W TDP, while the Max PRO 390 has a 55 W TDP.

Specification Differences

| Field | AMD Ryzen AI 9 365 | AMD Ryzen AI Max PRO 390 |

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

| Cores | 10 | 12 |

| Threads | 20 | 24 |

| Base Clock | 2.00 GHz | 3.20 GHz |

| Boost Clock | 5.00 GHz | 5.00 GHz |

| TDP | 28 W | 55 W |

| Socket | AMD Socket FP8 | AMD Socket FP11 |

| Codename | Strix Point | Strix Halo |

| Generation | Ryzen AI 300 (Zen 5 / Zen 5c) | Ryzen AI Max PRO (Zen 5) |

| L3 Cache | 16 MB | 64 MB (shared) |

| Memory Support | DDR5, LPDDR5X | LPDDR5X |

| Memory Bus | Dual-channel | Quad-channel |

| Memory Bandwidth | 89.6 GB/s | 256.0 GB/s |

| ECC Memory | No | Yes |

| Integrated Graphics | Radeon 880M | Radeon 8050S |

| Die Size | 233 mm² | Not recorded |

| Release Date | 2024-06-30 | 2025-01-05 |

| Part Number | 100-000001530 | 100-000001421 |

The Verdict

The benchmark data is unambiguous. The AMD Ryzen AI Max PRO 390 outperforms the AMD Ryzen AI 9 365 in 13 of 15 recorded tests, with margins ranging from 3.2% to 63.8%. The only tests where the 365 holds any advantage are single-core Cinebench results, and even there the lead is 0.8% or a tie. The Max PRO 390's higher core count (12 versus 10), higher base clock (3.20 GHz versus 2.00 GHz), four times the L3 cache (64 MB versus 16 MB), and nearly three times the memory bandwidth (256.0 GB/s versus 89.6 GB/s) all contribute to its dominance.

The average benchmark score tells the same story: 63,765 for the Max PRO 390 versus 40,048 for the 365. The Max PRO 390 sits in the 93rd percentile of all CPUs, while the 365 sits in the 87th. The nearest rival to the Max PRO 390 is the Intel Core i9-13900KS, a desktop flagship, with a 0.4% score difference. The 365's nearest rival is the AMD Ryzen 7 7700, a desktop mid-range part, with a 0.1% difference.

For workloads that stress multiple cores, high memory bandwidth, or large caches, the Max PRO 390 is the stronger part. For single-threaded tasks, the two are effectively equivalent, with the 365 holding a marginal edge in one Cinebench test. The 365's lower power envelope (28 W versus 55 W) may suit thinner mobile chassis, but the database does not include any thermal or battery measurements. On raw recorded performance, the Max PRO 390 is the clear winner, and the 365's only meaningful claim is single-core parity.

DETAILED SPECIFICATIONS

SPECIFICATION
AI 9 365
AI Max PRO 390
Core Specs
Cores
10
12 +20.0%
Threads
20
24 +20.0%
Base Clock (GHz)
2
3.2 +60.0%
Boost Clock (GHz)
5
5 0.0%
Frequency (GHz)
2
3.2 +60.0%
Turbo Clock (GHz)
5
5 0.0%
Multiplier
20
32 +60.0%
SMP CPUs
1
1 0.0%
Cache
L1 Cache
80 KB (per core)
80 KB (per core)
L2 Cache
1 MB (per core)
1 MB (per core)
L3 Cache
16 MB
64 MB (shared)
Power
TDP (W)
28
55 +96.4%
Configurable TDP
15-54 W
45-120 W
Architecture
Architecture
Zen 5
Zen 5
Codename
Strix Point
Strix Halo
Generation
Ryzen AI 300 (Zen 5 / Zen 5c)
Ryzen AI Max PRO (Zen 5)
Process Size
4 nm
4 nm
Die Size
233 mm²
—
Foundry
TSMC
TSMC
Memory
Memory Support
DDR5, LPDDR5X
LPDDR5X
Memory Bus
Dual-channel
Quad-channel
Memory Bandwidth
89.6 GB/s
256.0 GB/s
ECC Memory
No
Yes
Platform
Socket
AMD Socket FP8
AMD Socket FP11
PCIe
Gen 4, 16 Lanes(CPU only)
Gen 4, 16 Lanes(CPU only)
Intel Hybrid
Hybrid Cores
4 + 6
—
E-Core Frequency
1400 MHz up to 3.2 GHz
—
AI/NPU
NPU
Yes / 50 TOPS
Yes / 50 TOPS
Graphics
Integrated Graphics
Radeon 880M
Radeon 8050S
Other
Market
Mobile
Mobile
Production Status
Active
Active
Part Number
100-000001530
100-000001421
Package
FP8
FC-BGA
Tj Max
100°C
100°C
View Ryzen AI 9 365 Details View Ryzen AI Max PRO 390 Details