Intel Core Ultra 7 265K vs Intel Core Ultra 9 386H Comparison

Intel Core Ultra 7 265K and Intel Core Ultra 9 386H occupy different corners of the processor market, and the benchmark data reflects that split clearly. The Ultra 7 265K is a desktop part with a 125 TDP, while the Ultra 9 386H is a mobile chip with a 25 TDP. Across the 17 recorded head-to-head tests, the Ultra 7 265K wins 16, with the Ultra 9 386H taking only a single victory. The average benchmark score for the Ultra 7 265K is 70879 against 43210 for the Ultra 9 386H, placing the desktop chip in the 94th percentile of all CPUs versus the 88th percentile for the mobile part. This is not a close contest in raw throughput, but the mobile chip has specific strengths that matter for its intended use case.

Where Each One Wins

The Intel Core Ultra 7 265K wins every multi-core and most single-core workloads. Its largest margins come in memory-sensitive and instruction-heavy tasks: PassMark random string sorting shows an 89.3% lead, data compression is 88.9% ahead, and extended instructions are 86.5% higher. These are workloads that scale with core count, cache capacity, and memory bandwidth, all areas where the desktop part has a structural advantage.

The Intel Core Ultra 9 386H wins exactly one test: Cinebench R23 single-core. It scores 2071.5 versus 2020 for the Ultra 7 265K, a 2.5% margin. This is a narrow but real victory, suggesting that the Panther Lake architecture in the mobile chip has a slight per-core efficiency edge in lightly threaded rendering tasks. The mobile chip also shows competitive single-thread PassMark scores, 4218 versus 4928, a 16.8% deficit rather than the 50-80% gaps seen in multi-threaded tests.

For heavy parallel workloads, rendering, compression, encryption, and math-heavy simulation, the Ultra 7 265K is the clear choice. For short single-threaded bursts where power draw is a constraint, the Ultra 9 386H holds a small advantage. The data shows no scenario where the mobile chip wins multi-threaded work.

Architecture Differences

The two processors come from different Intel generations and foundries. The Ultra 7 265K uses Arrow Lake architecture, specifically Arrow Lake-S, built on a 3 nm process at TSMC. The Ultra 9 386H uses Panther Lake architecture, codename Panther Lake, also on a 3 nm process but fabricated at Intel. The desktop chip has 17,800 million transistors on a 243 mm² die, while the mobile chip has no recorded transistor or die size data.

Core configuration differs substantially. The Ultra 7 265K has 20 cores and 20 threads, meaning no hyperthreading. The Ultra 9 386H has 16 cores and 16 threads, also without hyperthreading. Cache hierarchies diverge: the Ultra 7 265K has 3 MB of L2 per core and 30 MB of shared L3, while the Ultra 9 386H has 2.5 MB of L2 per core and 18 MB of shared L3. Both have 192 KB of L1 per core.

Memory support differs. The Ultra 7 265K supports DDR5 only, with a dual-channel bus and 102.4 GB/s bandwidth. The Ultra 9 386H supports DDR5 and LPDDR5X, also dual-channel, but with higher recorded bandwidth at 115.2 GB/s. ECC memory is supported on the desktop chip but not on the mobile chip. PCIe lanes also differ: the Ultra 7 265K provides Gen 5 with 20 lanes, while the Ultra 9 386H provides Gen 5 with 12 lanes.

Integrated graphics are distinct. The Ultra 7 265K ships with Arc Xe-LPG Graphics with 64 execution units, while the Ultra 9 386H uses Intel Xe3 Graphics. The desktop chip has an unlocked multiplier; the mobile chip does not. The Ultra 7 265K uses Intel Socket 1851, the Ultra 9 386H uses Intel BGA 2540, reflecting the desktop versus mobile form factor split.

Head-to-Head Benchmarks

The largest single win for the Ultra 7 265K is in Cinebench R15 single-core, where it scores 708 against 303.5 for the Ultra 9 386H, a 133.3% lead. This is an outlier, likely reflecting different boost behavior in short-duration tests. In Cinebench R23 multi-core, the desktop chip scores 35850 versus 20547, a 74.5% advantage. Cinebench R20 multi-core shows 20918 versus 12820, a 63.2% lead, and R15 multi-core shows 5020 versus 3223, a 55.8% lead.

PassMark integer math shows 143242 versus 87284, a 64.1% lead for the desktop chip. Floating point math is 189629 versus 108527, a 74.7% lead. Data encryption shows 48246 versus 27150, a 77.7% gap. The desktop chip wins PassMark physics 3731 versus 3028, a smaller 23.2% margin, indicating the mobile chip is less disadvantaged in this specific test. PassMark multithread shows 58594 versus 35399, a 65.5% lead.

Single-threaded results are closer. PassMark single-thread shows 4928 versus 4218, a 16.8% lead for the desktop chip. The only mobile win, Cinebench R23 single-core, is 2071.5 versus 2020, a 2.5% margin. This indicates the Ultra 9 386H has competitive single-core capability but cannot sustain it across the broader PassMark suite.

Specification Differences

The core counts differ by four: 20 cores and 20 threads for the Ultra 7 265K, 16 cores and 16 threads for the Ultra 9 386H. Base clocks are 3.90 GHz versus 2.10 GHz, and boost clocks are 5.50 GHz versus 4.90 GHz, both favoring the desktop chip. TDP is 125 watts versus 25 watts, a fivefold difference that explains much of the performance gap.

Cache totals differ: the Ultra 7 265K has 30 MB of L3, the Ultra 9 386H has 18 MB. L2 per core is 3 MB versus 2.5 MB. Memory bandwidth is 102.4 GB/s versus 115.2 GB/s, a rare stat favoring the mobile chip. ECC support is present on the desktop chip only. PCIe lanes are 20 versus 12. The desktop chip has an unlocked multiplier, the mobile chip does not.

Release dates differ by over a year: the Ultra 7 265K launched on 2024-10-23, the Ultra 9 386H on 2026-01-04. The desktop chip has a launch MSRP of $394, while the mobile chip has no recorded launch MSRP. Part numbers are SRQCW for the desktop, SA4R5Q9EH for the mobile. The desktop chip is in the Core Ultra Series 2, the mobile chip in Core Ultra Series 3.

FAQ

Q: Which processor has more cores and threads?

A: The Intel Core Ultra 7 265K has 20 cores and 20 threads. The Intel Core Ultra 9 386H has 16 cores and 16 threads.

Q: Does the mobile chip ever beat the desktop chip in any benchmark?

A: Yes. The Intel Core Ultra 9 386H wins Cinebench R23 single-core with a score of 2071.5 versus 2020 for the Ultra 7 265K, a 2.5% margin.

Q: What is the difference in average benchmark scores?

A: The Ultra 7 265K has an average benchmark score of 70879, placing it in the 94th percentile. The Ultra 9 386H has an average score of 43210, placing it in the 88th percentile.

Q: Which processor supports ECC memory?

A: The Intel Core Ultra 7 265K supports ECC memory. The Intel Core Ultra 9 386H does not support ECC memory.

Q: How do the memory bandwidth figures compare?

A: The Ultra 7 265K has 102.4 GB/s bandwidth with DDR5 support. The Ultra 9 386H has 115.2 GB/s bandwidth with DDR5 and LPDDR5X support.

Q: Are both processors on the same manufacturing process?

A: Both are on a 3 nm process, but the Ultra 7 265K is fabricated at TSMC using Arrow Lake architecture, while the Ultra 9 386H is fabricated at Intel using Panther Lake architecture.

The Verdict

The data directs a clear split based on workload and platform. The Intel Core Ultra 7 265K is the superior processor for multi-threaded performance, winning 16 of 17 head-to-head tests with margins ranging from 16.8% in PassMark single-thread to 133.3% in Cinebench R15 single-core. Its 20 cores, 30 MB of L3 cache, and 5.50 GHz boost clock deliver decisive advantages in rendering, compression, encryption, and math workloads. The desktop chip also supports ECC memory and has an unlocked multiplier, making it suitable for systems where reliability and overclocking are priorities.

The Intel Core Ultra 9 386H is not a performance competitor to the desktop chip in any multi-threaded sense. Its single win, Cinebench R23 single-core at 2.5% higher, is narrow and isolated. The mobile chip does offer higher memory bandwidth at 115.2 GB/s and supports LPDDR5X, which is relevant for battery-constrained mobile systems. Its 25 watt TDP versus 125 watts indicates a fundamentally different power envelope, and the benchmark results show that this power cap costs roughly 40-90% in most multi-threaded tests.

For desktop users with sustained parallel workloads, the Ultra 7 265K is the only choice from this pair. For mobile users who need single-thread burst performance and low power draw, the Ultra 9 386H has a specific but limited edge. The recorded data does not support any scenario where the mobile chip outperforms the desktop chip in multi-threaded or mixed workloads. The 265K sits in the 94th percentile of all CPUs, the 386H in the 88th, and that gap is consistent with the head-to-head results.