AMD Ryzen AI 5 PRO 435G vs Intel Core Ultra 5 245T Comparison
AMD Ryzen AI 5 PRO 435G
Core Ultra 5 245T
PERFORMANCE BENCHMARKS
Analysis: AMD Ryzen AI 5 PRO 435G vs Intel Core Ultra 5 245T
The AMD Ryzen AI 5 PRO 435G and the Intel Core Ultra 5 245T are both active desktop processors with a 65W TDP, dual-channel DDR5 support, ECC memory, and locked multipliers. The database records an average score of 40718 for the AMD part and 38194 for the Intel part, with percentile ranks of 87 and 86 respectively. In the matched head-to-head PassMark set, however, the Intel part wins all 11 tests.
The Verdict
The matched benchmark data is lopsided. Intel wins 11 of 11 head-to-head PassMark tests, with recorded deltas from -10.7% in data compression to -83% in find prime numbers. The largest Intel margin is find prime numbers at -83%, followed by floating point math at -59.9%, physics at -56.1%, data encryption at -48.8%, and multithread at -34.2%. The AMD part's advantages in the database are aggregate ones. Its average benchmark score is 40718, above Intel's 38194, and its 87th percentile rank is above Intel's 86th. In the nearest-rival comparison, AMD sits 0.2% above the Intel Xeon 6357P (40630) and 0.3% above the Intel Core 5 223PE (40585), while Intel sits 0.1% below the AMD Ryzen 7 250 (38221) and 0.2% above the Intel Core i5-13600HX (38261). The data therefore supports Intel for any workload represented by the matched PassMark set, while the AMD part retains a higher aggregate database standing.
FAQ
Q: Which processor wins the recorded head-to-head benchmarks?
A: Intel wins all 11 matched PassMark tests. The recorded deltas range from -10.7% in data compression to -83% in find prime numbers. The head-to-head table records no AMD wins.
Q: How do core and thread counts differ?
A: AMD has 6 cores and 12 threads; Intel has 14 cores and 14 threads. AMD's thread count is double its core count, while Intel's thread count equals its core count.
Q: Which has the higher database average score?
A: AMD has an average score of 40718, while Intel has 38194. AMD also records an 87th percentile rank versus Intel's 86th. Despite this, Intel wins every matched head-to-head test.
Q: What are the cache differences?
A: AMD uses 80 KB of L1 per core, 1 MB of L2 per core, and 4 MB of L3. Intel uses 192 KB of L1 per core, 3 MB of L2 per core, and 24 MB of shared L3.
Q: What platform features differ?
A: AMD uses AMD Socket AM5, PCIe Gen 4 with 10 CPU lanes, and 89.6 GB/s of memory bandwidth. Intel uses Intel Socket 1851, PCIe Gen 5 with 20 CPU lanes, and 102.4 GB/s of memory bandwidth. Both support DDR5 dual-channel and ECC memory.
Architecture Differences
AMD's part uses the Gorgon Point codename, with the generation listed as Ryzen AI PRO 400 (Zen 5 / Zen 5c). Intel's part is Arrow Lake-S, with the architecture listed as Arrow Lake and the generation listed as Ultra 5 (Arrow Lake). Intel also has a series listing, Core Ultra Series 2, while AMD has no series listed. Both are fabricated by TSMC, but on different nodes: AMD uses 4 nm, Intel uses 3 nm. Intel's record includes 17,800 million transistors on a 243 mm² die; AMD's record does not list transistor or die-size figures.
The core and cache layouts differ. AMD has 6 cores and 12 threads; Intel has 14 cores and 14 threads. AMD's thread count is double its core count, while Intel's thread count equals its core count. AMD provides 80 KB of L1 and 1 MB of L2 per core plus 4 MB of L3. Intel provides 192 KB of L1 and 3 MB of L2 per core plus 24 MB of shared L3.
The platform architecture also differs. AMD uses AMD Socket AM5, PCIe Gen 4 with 10 CPU lanes, and 89.6 GB/s of memory bandwidth. Intel uses Intel Socket 1851, PCIe Gen 5 with 20 CPU lanes, and 102.4 GB/s of memory bandwidth. Both use DDR5 dual-channel memory and support ECC. The integrated graphics are Radeon 840M on AMD and Arc Xe-LPG Graphics 64EU on Intel.
Specification Differences
The table below lists the fields where the records differ. Both parts share a 65W TDP, DDR5 dual-channel memory support, ECC support, and locked multipliers.
| Field | AMD Ryzen AI 5 PRO 435G | Intel Core Ultra 5 245T |
| Series | Not listed | Core Ultra Series 2 |
| Architecture | Not listed | Arrow Lake |
| Codename | Gorgon Point | Arrow Lake-S |
| Generation | Ryzen AI PRO 400 (Zen 5 / Zen 5c) | Ultra 5 (Arrow Lake) |
| Process node | 4 nm | 3 nm |
| Cores | 6 | 14 |
| Threads | 12 | 14 |
| Base clock | 2.00 GHz | 2.20 GHz |
| Boost clock | 4.50 GHz | 5.10 GHz |
| Socket | AMD Socket AM5 | Intel Socket 1851 |
| L1 cache | 80 KB per core | 192 KB per core |
| L2 cache | 1 MB per core | 3 MB per core |
| L3 cache | 4 MB | 24 MB shared |
| Memory bandwidth | 89.6 GB/s | 102.4 GB/s |
| PCIe | Gen 4, 10 Lanes (CPU only) | Gen 5, 20 Lanes (CPU only) |
| Integrated graphics | Radeon 840M | Arc Xe-LPG Graphics 64EU |
| Transistors | Not listed | 17,800 million |
| Die size | Not listed | 243 mm² |
| Release date | 2026-03-01 | 2025-01-06 |
| Launch MSRP | Not listed | $270 |
| Part number | 100-000001783 | SRVFF |
The specification split aligns with the benchmark pattern. Intel has more cores, higher base and boost clocks, a larger L3, and higher memory bandwidth. Intel also uses a 3 nm node, while AMD uses 4 nm. The sockets are different, so platform compatibility is not shared.
Head-to-Head Benchmarks
Intel wins every entry in the matched head-to-head table. The scores and recorded deltas are as follows:
| Test | AMD | Intel | Delta |
| data_compression | 253484 | 283812 | -10.7% |
| data_encryption | 12111 | 23656 | -48.8% |
| extended_instructions | 18697 | 22071 | -15.3% |
| find_prime_numbers | 55 | 324 | -83% |
| floating_point_math | 43494 | 108499 | -59.9% |
| integer_math | 63707 | 88676 | -28.2% |
| multithread | 20285 | 30833 | -34.2% |
| physics | 999 | 2278 | -56.1% |
| random_string_sorting | 27407 | 34931 | -21.5% |
| single_thread | 3829 | 4367 | -12.3% |
| singlethread | 3829 | 4367 | -12.3% |
The largest Intel advantage is in find_prime_numbers: AMD scores 55, Intel scores 324, and the recorded delta is -83%. Floating point math is also heavily Intel-favored at 43494 vs 108499, -59.9%. Physics follows at 999 vs 2278, -56.1%, and data encryption at 12111 vs 23656, -48.8%. Multithread shows 20285 vs 30833, -34.2%, and integer math shows 63707 vs 88676, -28.2%. Random string sorting is 27407 vs 34931, -21.5%, and extended instructions is 18697 vs 22071, -15.3%. The narrowest margin is data compression at 253484 vs 283812, -10.7%, followed by single-thread at 3829 vs 4367, -12.3%. The database records the single-thread test twice with identical scores.
Where Each One Wins
Based on the matched head-to-head set, the Intel Core Ultra 5 245T wins every workload category. The list includes integer math, floating point math, data encryption, data compression, extended instructions, find prime numbers, random string sorting, physics, multithread, and single-thread. There is no recorded matched test where the AMD Ryzen AI 5 PRO 435G takes the lead.
The AMD part does hold a higher aggregate position. Its average score of 40718 exceeds Intel's 38194, and its 87th percentile is above Intel's 86th. In nearest-rival terms, AMD is 0.2% ahead of the Intel Xeon 6357P (40630) and 0.3% ahead of the Intel Core 5 223PE (40585); Intel is 0.1% behind the AMD Ryzen 7 250 (38221) and 0.2% ahead of the Intel Core i5-13600HX (38261). The AMD nearest-rival list also includes Intel Core 7 253PE (40557, delta 0.4%) and Intel Core Ultra X7 368H (40518, delta 0.5%). The Intel nearest-rival list also includes Intel Core i5-14490F (38149, delta 0.1%) and Intel Core i5-13600KF (38103, delta 0.2%). This split means the Intel part dominates the matched PassMark workloads, while the AMD part ranks higher in the database's overall CPU population.
The specification data aligns with the matched results. Intel's 14 cores, 5.10 GHz boost clock, 24 MB shared L3, and 102.4 GB/s memory bandwidth are all reflected in its PassMark wins. AMD's 6 cores, 4.50 GHz boost clock, 4 MB L3, and 89.6 GB/s memory bandwidth leave it behind in every recorded head-to-head test. The AMD part's higher average score comes from its own recorded benchmark list, which does not include the Cinebench scores recorded for Intel.