AMD Ryzen AI 9 PRO 465 vs Intel Arc G3 EXTREME Comparison

Where Each One Wins

The benchmark split between the AMD Ryzen AI 9 PRO 465 and the Intel Arc G3 EXTREME is nearly even, with Intel taking 6 of the 11 head-to-head tests and AMD taking 5. The decisive pattern is not about total wins but about workload type: AMD dominates integer-heavy and data-processing tasks, while Intel leads in floating-point, physics, and single-thread scenarios.

AMD's largest margins come from integer math, where it scores 107173 against Intel's 71164, a 50.6% advantage. Data compression shows a 25.4% lead (385174 vs 307094), and extended instructions deliver a 10.1% edge (26441 vs 24017). Random string sorting adds another win at 40860 vs 37331, a 9.5% difference. The multithread test is closer, with AMD at 31485 and Intel at 30659, only a 2.7% gap, but it still goes to AMD.

Intel's wins are equally decisive in the other direction. Floating-point math shows 86929 vs 66824, a 23.1% lead. Physics simulation favors Intel at 2515 vs 1747, a 30.5% margin. Find prime numbers is Intel's biggest win, 248 vs 126, a 49.2% difference. Data encryption goes to Intel (23881 vs 19308, 19.1% ahead), and single-thread performance shows Intel at 4293 vs 4168, a 2.9% lead.

The overall average benchmark score tells a different story: AMD's average is 62498, while Intel's is 36699. That gap reflects the different benchmark sets each CPU was tested with, not raw performance parity. AMD sits at the 93rd percentile of all CPUs, while Intel is at the 85th percentile. AMD's nearest rivals include the Intel Core Ultra 7 255HX (0.4% below) and the AMD Ryzen AI Embedded P185 (0.5% below), while Intel's closest competitor is the Intel Core Ultra 5 225 (0.6% below it).

Architecture Differences

The two processors come from different design philosophies. AMD uses Zen 5 architecture on a 4 nm TSMC process, codenamed Gorgon Point, part of the Ryzen AI PRO 400 generation with Zen 5 / Zen 5c cores. Intel uses Panther Lake on a 3 nm Intel process, belonging to the Arc G3 generation.

Core counts differ significantly. AMD has 10 cores and 20 threads, which means simultaneous multithreading is enabled. Intel has 14 cores but only 14 threads, indicating no SMT on those cores. This explains why AMD's multithread score (31485) edges out Intel's (30659) despite Intel having more physical cores.

Cache layouts are also distinct. 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, 2.5 MB of L2 per core, and 18 MB of shared L3. Intel's larger per-core L1 and L2 caches support its single-thread and floating-point strengths, while AMD's smaller but sufficient caches feed its integer throughput.

Clock speeds are close: AMD's base is 2.00 GHz with a 5.00 GHz boost, while Intel's base is 1.90 GHz with a 4.70 GHz boost. AMD holds a slight clock advantage, but Intel's architectural efficiency in certain workloads compensates.

Memory support differs. AMD supports DDR5 and LPDDR5X with dual-channel memory and 89.6 GB/s bandwidth. Intel supports only LPDDR5X, also dual-channel, but with 136.5 GB/s bandwidth, a 52% higher figure. This memory bandwidth advantage likely contributes to Intel's physics and floating-point results.

PCIe generations diverge: AMD uses Gen 4 with 16 CPU lanes, while Intel uses Gen 5 with 4 CPU lanes. Integrated graphics also differ: AMD has Radeon 890M, Intel has Arc B390. Both use mobile sockets (AMD Socket FP8 for AMD, Intel BGA 2540 for Intel) and both are active production parts. Neither supports ECC memory, and neither has an unlocked multiplier.

Head-to-Head Benchmarks

The most striking result is integer math, where AMD's 107173 crushes Intel's 71164 by 50.6%. This is the single largest delta in either direction, and it shows AMD's Zen 5 cores with SMT handle integer workloads exceptionally well. Data compression follows a similar pattern: AMD at 385174 vs Intel at 307094, a 25.4% lead. Extended instructions show AMD at 26441 vs 24017, a 10.1% gap, and random string sorting adds another AMD win at 40860 vs 37331, 9.5% ahead.

Intel's counterattack is led by prime number finding, where 248 beats 126 by 49.2%. Physics simulation shows Intel at 2515 vs 1747, a 30.5% margin. Floating-point math gives Intel a 23.1% lead (86929 vs 66824), and data encryption shows 23881 vs 19308, a 19.1% edge. Single-thread performance is a narrow Intel win: 4293 vs 4168, just 2.9% apart.

The multithread test is the closest of all, with AMD at 31485 and Intel at 30659. AMD wins by 2.7%, but the margin is small enough that real-world multithreaded applications could see either processor ahead depending on thread scheduling and workload mix. The fact that AMD wins this test with 10 cores and 20 threads against Intel's 14 cores and 14 threads suggests AMD's SMT implementation is effective.

The data also shows a clear pattern in the benchmark averages. AMD's average score of 62498 places it among desktop-class rivals like the Intel Core i9-13900KF (1.1% above AMD) and the Intel Core i7-13790F (0.9% below). Intel's average of 36699 places it alongside the AMD Ryzen 5 5600F (0.7% below) and the AMD Ryzen 5 5500X3D (0.9% below). This suggests AMD's overall benchmark profile is more competitive with high-end desktop parts, while Intel's profile aligns with mid-range desktop CPUs, despite both being mobile processors.

FAQ

Q: Which processor has better single-thread performance?

A: The Intel Arc G3 EXTREME leads in the PassMark single-thread test with 4293 points versus AMD's 4168, a 2.9% difference. Intel's higher per-core cache (192 KB L1 and 2.5 MB L2 per core) likely supports this result.

Q: How do they compare in multithreaded workloads?

A: AMD wins the PassMark multithread test with 31485 points against Intel's 30659, a 2.7% margin. AMD achieves this with 10 cores and 20 threads, while Intel has 14 cores and 14 threads.

Q: Which processor has the larger memory bandwidth?

A: Intel has 136.5 GB/s versus AMD's 89.6 GB/s, a 52% advantage. Intel supports only LPDDR5X memory, while AMD supports both DDR5 and LPDDR5X.

Q: What is the biggest benchmark win for each processor?

A: AMD's largest win is integer math, scoring 107173 against Intel's 71164, a 50.6% lead. Intel's largest win is find prime numbers, scoring 248 against AMD's 126, a 49.2% lead.

Q: How do the integrated graphics differ?

A: AMD uses Radeon 890M graphics, while Intel uses Arc B390 graphics. Both are integrated into mobile processors, but the database does not include benchmark results for these GPU units.

Q: Which processor has the higher average benchmark score?

A: AMD has an average benchmark score of 62498, while Intel has 36699. AMD's percentile rank is 93 versus Intel's 85, though the different benchmark sets used for each processor must be considered when interpreting this gap.

The Verdict

The data supports a clear workload-based choice. The AMD Ryzen AI 9 PRO 465 is the stronger processor for integer-heavy tasks, data compression, extended instruction workloads, and random string sorting. Its 50.6% lead in integer math and 25.4% lead in data compression make it the preferred option for database operations, cryptographic hashing, and general-purpose compute that relies on integer arithmetic. The multithread win, however narrow, also favors AMD for parallel workloads that can use SMT effectively.

The Intel Arc G3 EXTREME is the better choice for floating-point math, physics simulation, prime number finding, and data encryption. Its 23.1% floating-point lead and 30.5% physics advantage suggest it handles scientific computing, simulation, and encryption workloads more efficiently. The single-thread win, even at 2.9%, gives Intel a slight edge in lightly threaded applications. The significantly higher memory bandwidth (136.5 GB/s vs 89.6 GB/s) also positions Intel well for bandwidth-sensitive tasks.

Looking at overall competitive standing, AMD's average score of 62498 places it near the Intel Core Ultra 7 255HX (0.4% below) and the AMD Ryzen AI Embedded P185 (0.5% below). Intel's average of 36699 sits near the Intel Core Ultra 5 225 (0.6% below) and the AMD Ryzen 5 5600F (0.7% below). This places AMD in a higher performance tier overall, even though Intel wins more individual head-to-head tests. The 93rd versus 85th percentile ranking reinforces this interpretation.

For users who prioritize integer throughput, data compression, and multithreaded server-style workloads, the AMD Ryzen AI 9 PRO 465 is the justified pick. For users who need floating-point performance, physics simulation, encryption speed, and the best single-thread results, the Intel Arc G3 EXTREME delivers those specific strengths. The choice is not about which is universally faster, but which set of workload advantages matches the intended use case.

Specification Differences

| Specification | AMD Ryzen AI 9 PRO 465 | Intel Arc G3 EXTREME |

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

| Cores | 10 | 14 |

| Threads | 20 | 14 |

| Base Clock | 2.00 GHz | 1.90 GHz |

| Boost Clock | 5.00 GHz | 4.70 GHz |

| TDP | 28 W | 25 W |

| Socket | AMD Socket FP8 | Intel BGA 2540 |

| Architecture | Zen 5 | Not specified |

| Codename | Gorgon Point | Panther Lake |

| Process Node | 4 nm | 3 nm |

| Foundry | TSMC | Intel |

| Die Size | 233 mm² | Not specified |

| L1 Cache | 80 KB per core | 192 KB per core |

| L2 Cache | 1 MB per core | 2.5 MB per core |

| L3 Cache | 16 MB | 18 MB shared |

| Memory Support | DDR5, LPDDR5X | LPDDR5X |

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

| Memory Bandwidth | 89.6 GB/s | 136.5 GB/s |

| ECC Memory | No | No |

| PCIe | Gen 4, 16 lanes | Gen 5, 4 lanes |

| Integrated Graphics | Radeon 890M | Arc B390 |

| Market Segment | Mobile | Mobile |

| Production Status | Active | Active |

| Release Date | 2026-01-04 | 2026-05-27 |

| Multiplier Unlocked | No | No |

| Part Number | 100-000001862 | SA4R3 |