AMD Ryzen AI 7 345 vs Intel Core Ultra 9 288V Comparison

The Verdict

The benchmark data splits these two mobile processors almost evenly, with Intel taking 8 of 15 head-to-head tests and AMD taking 7. The AMD Ryzen AI 7 345 wins the heavily multithreaded Cinebench tests by double digits, while the Intel Core Ultra 9 288V dominates integer math, floating-point math, and physics workloads. The AMD part sits at the 81st percentile of all CPUs in the database with an average benchmark score of 29461, while the Intel part sits at the 76th percentile with an average score of 23219. That percentile gap indicates AMD has the broader overall performance profile. Buyers who prioritize heavily parallel rendering workloads should pick the AMD Ryzen AI 7 345. Buyers whose workloads lean on single-thread responsiveness, cryptographic operations, or physics simulation should pick the Intel Core Ultra 9 288V. The Intel chip also carries a 33.5% advantage in PassMark physics, which matters for simulation tasks. The AMD chip counters with a 44.2% advantage in integer math, which matters for general productivity and database work.

Architecture Differences

The two processors come from different process nodes and design philosophies. The AMD Ryzen AI 7 345 uses TSMC's 4 nm process with a hybrid Zen 5 / Zen 5c core arrangement under the Krackan Point codename. It has 6 cores and 12 threads, relying on simultaneous multithreading to fill its execution slots. The Intel Core Ultra 9 288V uses TSMC's 3 nm process under the Lunar Lake codename and belongs to the Core Ultra Series 2. It has 8 cores and 8 threads, with no multithreading. This core count difference matters: Intel has more physical cores, but AMD has more logical threads. The Intel part also has a substantially larger L3 cache at 12 MB shared versus AMD's 4 MB, and Intel's per-core L1 and L2 allocations are larger as well. Intel specifies 192 KB of L1 per core and 2.5 MB of L2 per core, while AMD specifies 80 KB of L1 per core and 1 MB of L2 per core.

Memory architecture differs too. AMD supports both DDR5 and LPDDR5X through a dual-channel bus with 89.6 GB/s of bandwidth. Intel supports only LPDDR5X, also dual-channel, but with 136.5 GB/s of bandwidth. That is a 52.3% bandwidth advantage for Intel, which helps explain its strong showing in memory-sensitive workloads like physics and floating-point math. PCIe support differs as well: AMD provides Gen 4 with 14 CPU lanes, while Intel provides Gen 5 with 4 CPU lanes. The integrated graphics also differ, with AMD using the Radeon 840M and Intel using the Arc 140V. Both chips are unlocked in the sense that neither has an unlocked multiplier, and both are active production parts for the mobile segment. Intel reached the market earlier, with a release date of 2024-09-23, while AMD's release date is 2025-01-14.

FAQ

Q: Which processor has the higher average benchmark score?

A: The AMD Ryzen AI 7 345 has an average benchmark score of 29461, compared to 23219 for the Intel Core Ultra 9 288V. AMD also sits at the 81st percentile of all CPUs, while Intel sits at the 76th percentile.

Q: Which chip wins in single-core performance?

A: The Intel Core Ultra 9 288V wins both recorded single-core tests. It scores 301.5 in Cinebench R15 single-core versus 271 for AMD, a 10.1% margin, and 1950 in Cinebench R23 single-core versus 1818, a 6.8% margin. In PassMark single-thread, Intel scores 4274 versus 3875, a 9.3% margin.

Q: Which chip wins in multithreaded rendering?

A: The AMD Ryzen AI 7 345 wins both Cinebench multicore tests. It scores 1712 in Cinebench R15 multicore versus 1583 for Intel, an 8.1% margin, and 11461 in Cinebench R23 multicore versus 10178, a 12.6% margin.

Q: How do the two compare in integer math?

A: The AMD Ryzen AI 7 345 delivers 63475 in PassMark integer math versus 44019 for Intel, a 44.2% advantage. This is the largest win in either direction across all 15 head-to-head tests.

Q: How do the two compare in floating-point math?

A: The Intel Core Ultra 9 288V delivers 59536 in PassMark floating-point math versus 42621 for AMD, a 28.4% advantage.

Q: What are the core and thread counts for each processor?

A: The AMD Ryzen AI 7 345 has 6 cores and 12 threads. The Intel Core Ultra 9 288V has 8 cores and 8 threads.

Specification Differences

The two processors differ across nearly every core specification. AMD uses 6 cores and 12 threads, while Intel uses 8 cores and 8 threads. AMD's base clock is 2.00 GHz with a boost clock of 4.60 GHz. Intel's base clock is 3.30 GHz with a boost clock of 5.10 GHz. Intel holds the frequency advantage at both the base and boost levels. Thermal design power differs slightly: AMD is rated at 28 W and Intel at 30 W. The sockets are incompatible: AMD uses AMD Socket FP8, while Intel uses Intel BGA 2833.

Cache configurations differ substantially. AMD provides 80 KB of L1 per core, 1 MB of L2 per core, and 4 MB of L3. Intel provides 192 KB of L1 per core, 2.5 MB of L2 per core, and 12 MB of shared L3. Intel's L3 allocation is three times larger than AMD's. Memory support differs as well: AMD supports DDR5 and LPDDR5X, while Intel supports only LPDDR5X. Both use dual-channel memory buses, but Intel's bandwidth is 136.5 GB/s versus 89.6 GB/s for AMD. PCIe support differs in both generation and lane count: AMD offers Gen 4 with 14 CPU lanes, while Intel offers Gen 5 with 4 CPU lanes. Neither supports ECC memory. The integrated GPUs are different: AMD uses the Radeon 840M, Intel uses the Arc 140V. Process nodes differ, with AMD on TSMC 4 nm and Intel on TSMC 3 nm. Neither processor has an unlocked multiplier. The codenames differ, with AMD using Krackan Point and Intel using Lunar Lake, as does the generation label, with AMD listed as Ryzen AI 300 (Zen 5 / Zen 5c) and Intel as Ultra 9 (Lunar Lake).

Head-to-Head Benchmarks

The AMD Ryzen AI 7 345 takes the multithreaded rendering crown. In Cinebench R23 multicore, AMD scores 11461 against Intel's 10178, a 12.6% margin. In Cinebench R15 multicore, AMD scores 1712 against Intel's 1583, an 8.1% margin. These wins reflect AMD's 12 threads against Intel's 8, even though Intel has two more physical cores. The AMD chip also wins PassMark multithread, but narrowly: 19927 versus 19810, a 0.6% margin. This near-tie shows that the two chips deliver similar aggregate throughput despite their architectural differences.

The integer math test is the biggest single gap in the entire comparison. AMD scores 63475 in PassMark integer math, beating Intel's 44019 by 44.2%. Data compression also favors AMD heavily: 237484 versus 186521, a 27.3% margin. Extended instructions go to AMD at 17003 versus 15613, an 8.9% margin, and random string sorting goes to AMD at 25435 versus 22622, a 12.4% margin. These results indicate AMD's core design excels at branch-heavy, integer-oriented workloads.

The Intel Core Ultra 9 288V takes the single-core tests. Cinebench R15 single-core goes to Intel at 301.5 versus 271, a 10.1% margin. Cinebench R23 single-core goes to Intel at 1950 versus 1818, a 6.8% margin. PassMark single-thread goes to Intel at 4274 versus 3875, a 9.3% margin. Intel's higher boost clock of 5.10 GHz against AMD's 4.60 GHz supports this pattern.

The physics test shows the largest Intel win after integer math. Intel scores 1637 in PassMark physics versus 1089 for AMD, a 33.5% margin. Floating-point math also favors Intel: 59536 versus 42621, a 28.4% margin. Prime number finding heavily favors Intel: 195 versus 62, a 68.2% margin, which is the largest percentage gap in the entire set. Data encryption goes to Intel at 14141 versus 11814, a 16.5% margin.

Where Each One Wins

The AMD Ryzen AI 7 345 wins in workloads that scale with thread count and integer throughput. Cinebench multicore results show a 12.6% advantage in R23 and an 8.1% advantage in R15, making AMD the better choice for rendering and video encoding workloads. The 44.2% integer math advantage and the 27.3% data compression advantage point to strong performance in compilation, spreadsheet calculations, and database operations. The 12.4% random string sorting win reinforces this pattern. Extended instructions at 8.9% ahead give AMD a modest edge in vectorized general-purpose code. The PassMark multithread result, while only 0.6% ahead, confirms that AMD maintains the throughput lead even in a mixed workload.

The Intel Core Ultra 9 288V wins in single-thread-sensitive and physics-heavy workloads. The 10.1% and 6.8% Cinebench single-core margins, plus the 9.3% PassMark single-thread margin, make Intel the stronger choice for lightly threaded applications like office document handling and web browsing. The 33.5% physics advantage and the 28.4% floating-point math advantage indicate a strong showing in scientific computing and simulation workloads. The 68.2% prime number finding advantage is exceptional and points to superior performance in workloads with tight loops and simple integer operations. The 16.5% data encryption advantage makes Intel the better option for cryptographic tasks and secure communication workloads. The 136.5 GB/s memory bandwidth versus 89.6 GB/s for AMD likely contributes to these wins in memory-intensive tests.

The final tally of 7 wins for AMD and 8 wins for Intel masks the magnitude differences. AMD's wins include the two largest margins in the entire comparison, with integer math at 44.2% and data compression at 27.3%. Intel's wins include the single largest percentage margin, with prime number finding at 68.2%, but several of Intel's wins are in the 6% to 16% range. The average benchmark score of 29461 for AMD versus 23219 for Intel gives AMD a 26.9% overall advantage, and the 5-point percentile gap (81 versus 76) reinforces that AMD is the stronger all-around performer in the database's aggregate metrics.