AMD Ryzen AI 7 345 vs Intel Core Ultra 7 266V Comparison

Head-to-Head Benchmarks

The head-to-head results split nearly evenly, with Intel taking 8 of 15 recorded tests and AMD taking 7, but the margins tell a more complex story than the win count. The Intel Core Ultra 7 266V dominates in multi-core rendering and several compute-heavy workloads, while the AMD Ryzen AI 7 345 posts its largest victories in integer math and data compression.

The single largest delta comes from PassMark integer math, where the AMD Ryzen AI 7 345 scores 63,475 against 41,558 for the Intel part, a 52.7% advantage. Data compression similarly favors AMD by 27%, with scores of 237,484 versus 187,050. Random string sorting adds another AMD win at 11% (25,435 versus 22,905), and extended instructions favor AMD by 6.7% (17,003 versus 15,928). The Cinebench R15 multi-core test goes to AMD by a narrow 2.7% margin (1,712 versus 1,667), and the R15 single-core test shows AMD ahead by 15.3% (271 versus 235). PassMark multithread also lands on the AMD side, but only by 2.4% (19,927 versus 19,461).

The Intel Core Ultra 7 266V answers with the largest overall swing in the entire comparison: Cinebench R23 multi-core. Here Intel scores 16,544 against AMD's 11,461, a 30.7% deficit for AMD. Cinebench R23 single-core goes Intel's way by 22.1% (2,335 versus 1,818). PassMark physics shows Intel ahead by 32.3% (1,608 versus 1,089), and floating point math favors Intel by 25.1% (56,923 versus 42,621). Prime number finding is a crushing Intel victory at 191 versus 62, a 67.5% gap, the second-largest delta in the dataset. Data encryption goes to Intel by 14.5% (13,822 versus 11,814), and PassMark single-thread is essentially a tie: Intel leads by 1.7% (3,943 versus 3,875).

The average benchmark scores in the database place the AMD part at 29,461 overall, versus 23,297 for Intel. AMD's nearest rivals include the AMD Ryzen 7 3800X at 29,447 (0% delta), the AMD EPYC 9654 at 29,409 (0.2% delta), and the Intel Core i5-13500HX at 29,270 (0.7% delta). Intel's nearest rivals include the AMD Ryzen 7 5800H at 23,277 (0.1% delta), the Intel Core i9-11900F at 23,254 (0.2% delta), and the Intel Core Ultra 9 288V at 23,219 (0.3% delta). The percentile ranks confirm the split: AMD sits at the 81st percentile among all CPUs, while Intel sits at the 76th.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD Ryzen AI 7 345 uses the Krackan Point codename, belongs to the Ryzen AI 300 generation built on Zen 5 / Zen 5c cores, and is manufactured on a 4 nm process at TSMC. The Intel Core Ultra 7 266V belongs to the Core Ultra Series 2, uses the Lunar Lake architecture, and is manufactured on a 3 nm process, also at TSMC.

Core configuration differs sharply. AMD provides 6 cores and 12 threads, meaning simultaneous multithreading is enabled. Intel provides 8 cores and 8 threads, with no multithreading. The clock profiles also diverge: AMD lists a 2.00 GHz base and 4.60 GHz boost, while Intel lists a 2.20 GHz base and 5.00 GHz boost. Power envelopes are reversed in magnitude: AMD has a 28 W TDP, Intel has a 17 W TDP.

Cache structures are not comparable in size or organization. AMD allocates 80 KB of L1 per core, 1 MB of L2 per core, and 4 MB of L3. Intel allocates 192 KB of L1 per core, 2.5 MB of L2 per core, and 12 MB of shared L3. The L3 difference is the most striking architectural gap: Intel's shared 12 MB pool is three times larger than AMD's 4 MB allocation.

Memory support shows both use dual-channel buses, but the bandwidth figures differ. AMD lists 89.6 GB/s with DDR5 and LPDDR5X support. Intel lists 136.5 GB/s with LPDDR5X support that depends on the motherboard. Neither supports ECC memory.

PCIe connectivity is another split. AMD provides Gen 4 with 14 lanes (CPU only). Intel provides Gen 5 with 4 lanes (CPU only). The integrated graphics also differ: AMD pairs the CPU with the Radeon 840M, while Intel pairs it with the Arc 140V.

Sockets are incompatible: AMD uses AMD Socket FP8, Intel uses Intel BGA 2833. Release dates place Intel first at 2024-09-23, with AMD following on 2025-01-14. Both parts are marked Active in production status, and neither has an unlocked multiplier.

Where Each One Wins

The AMD Ryzen AI 7 345 wins in workloads that reward high bandwidth integer operations and data throughput. Integer math, data compression, random string sorting, and extended instruction execution all fall on the AMD side. The 52.7% integer math lead and the 27% compression lead indicate strong general-purpose compute for office-style tasks, database work, and file handling. The Cinebench R15 results, both single and multi-core, also favor AMD, suggesting that older rendering benchmarks respond better to AMD's core configuration. The multithread result, though narrow (2.4%), further supports AMD's advantage in tasks that scale with threads rather than raw frequency.

The Intel Core Ultra 7 266V wins in scenarios that stress modern multi-core rendering and single-core frequency. The Cinebench R23 multi-core score is 30.7% ahead, which is a decisive margin for current-generation rendering workloads. The R23 single-core score is 22.1% ahead, showing a strong frequency advantage at 5.00 GHz boost. Physics simulation goes Intel's way by 32.3%, and floating point math by 25.1%. Prime number finding, a classic integer-heavy but scalar workload, sees Intel ahead by 67.5%. Data encryption also lands on Intel, up 14.5%. The PassMark single-thread score, while only 1.7% ahead, still tips to Intel.

The pattern is consistent: AMD leads in parallel integer throughput and data manipulation; Intel leads in scalar floating point, physics, and the newest Cinebench rendering tests. The average benchmark score gap (29,461 versus 23,297) favors AMD overall, but that aggregate hides the fact that Intel wins the most recent rendering benchmark by a wide margin.

FAQ

Q: Which processor has the higher Cinebench R23 multi-core score?

A: The Intel Core Ultra 7 266V scores 16,544 versus 11,461 for the AMD Ryzen AI 7 345, a 30.7% lead.

Q: Where does the AMD Ryzen AI 7 345 have its largest benchmark advantage?

A: In PassMark integer math, AMD scores 63,475 against 41,558, a 52.7% advantage.

Q: What is the core and thread count difference?

A: The AMD Ryzen AI 7 345 has 6 cores and 12 threads. The Intel Core Ultra 7 266V has 8 cores and 8 threads.

Q: How do the TDP values compare?

A: The AMD part has a 28 W TDP, while the Intel part has a 17 W TDP.

Q: Which processor has the higher boost clock?

A: The Intel Core Ultra 7 266V boosts to 5.00 GHz, while the AMD Ryzen AI 7 345 boosts to 4.60 GHz.

Q: What are the process nodes for each processor?

A: The AMD Ryzen AI 7 345 uses a 4 nm process, and the Intel Core Ultra 7 266V uses a 3 nm process. Both are fabricated by TSMC.

Specification Differences

| Specification | AMD Ryzen AI 7 345 | Intel Core Ultra 7 266V |

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

| Cores | 6 | 8 |

| Threads | 12 | 8 |

| Base clock | 2.00 GHz | 2.20 GHz |

| Boost clock | 4.60 GHz | 5.00 GHz |

| TDP | 28 W | 17 W |

| Socket | AMD Socket FP8 | Intel BGA 2833 |

| Codename | Krackan Point | Lunar Lake |

| Generation | Ryzen AI 300 (Zen 5 / Zen 5c) | Ultra 7 (Lunar Lake) |

| Process node | 4 nm | 3 nm |

| L1 cache | 80 KB (per core) | 192 KB (per core) |

| L2 cache | 1 MB (per core) | 2.5 MB (per core) |

| L3 cache | 4 MB | 12 MB (shared) |

| Memory bandwidth | 89.6 GB/s | 136.5 GB/s |

| PCIe | Gen 4, 14 Lanes (CPU only) | Gen 5, 4 Lanes (CPU only) |

| Integrated graphics | Radeon 840M | Arc 140V |

| Release date | 2025-01-14 | 2024-09-23 |

| Part number | 100-000002122 | SRPMMSRPMY |

The Verdict

The recorded data supports a clear split based on workload type and power budget. The AMD Ryzen AI 7 345 is the stronger choice for integer-heavy data processing, compression workloads, and tasks that benefit from 12 threads. Its 52.7% integer math lead, 27% compression lead, and 2.4% multithread win all point to broad parallel throughput. The 81st percentile ranking among all CPUs also places it higher than Intel's 76th percentile.

The Intel Core Ultra 7 266V is the stronger choice for current-generation rendering and scalar floating point workloads. Its 30.7% Cinebench R23 multi-core lead and 22.1% single-core lead are substantial, and the 67.5% prime number finding advantage shows extreme efficiency in a specific scalar workload. The 17 W TDP, lower than AMD's 28 W, also indicates a more power-efficient design on a smaller 3 nm process.

The overall average benchmark score favors AMD at 29,461 versus 23,297, but the largest single benchmark delta favors Intel. The choice depends on whether the target workload resembles Cinebench R23 and physics simulation, where Intel leads decisively, or integer math and data compression, where AMD leads by even larger margins. The data does not declare a global winner; it declares two domain-specific winners.