Intel Core Ultra 7 265F vs Intel Core Ultra X7 358H Comparison

Where Each One Wins

The recorded benchmark data presents an unambiguous picture: the Intel Core Ultra 7 265F wins all 17 head-to-head comparisons against the Intel Core Ultra X7 358H. The desktop processor takes every single test across Cinebench R15, R20, R23, and the Passmark suite, with zero wins recorded for the mobile chip.

The largest margins appear in multi-core rendering workloads. In Cinebench R23 multi-core, the Ultra 7 265F scores 41980 against 18747 for the Ultra X7 358H, a 123.9% advantage. Single-core R23 shows an even wider gap at 184.9%, with scores of 5926 and 2080 respectively. These are not incremental differences; they represent fundamentally different performance tiers.

The desktop part also dominates memory-sensitive and instruction-heavy tasks. Passmark data compression shows 507018 versus 332508, a 52.5% lead. Floating point math delivers 173855 against 103842, a 67.4% gap. Integer math follows at 66.1% ahead with scores of 138078 and 83147. Extended instructions testing shows a 43.9% edge (39235 versus 27274).

The narrowest margin appears in Passmark physics, where the Ultra 7 265F scores 3172 against 3021, just 5% ahead. This suggests the mobile part retains competitive physics simulation capability despite its lower core count and thermal envelope. Single-thread Passmark results are also relatively close at 15.2%, with 4750 versus 4124.

For users prioritizing raw throughput, the 265F is the clear choice. The 358H remains competitive in lighter single-thread workloads, but the data shows no scenario where it outperforms the desktop chip.

Architecture Differences

The two processors come from different Intel families and target entirely different market segments. The Core Ultra 7 265F belongs to the Core Ultra Series 2, uses the Arrow Lake architecture (Arrow Lake-S codename), and is a desktop part on Intel Socket 1851. The Core Ultra X7 358H is from Core Ultra Series 3, uses Panther Lake (Panther Lake-H codename), and is a mobile processor on Intel BGA 2540.

Both are built on a 3 nm process, but the foundries differ. The 265F is manufactured by TSMC, while the 358H uses Intel's own fabs. The desktop chip packs 17,800 million transistors on a 243 mm² die, while transistor count and die size for the mobile part are not recorded in the database.

Core and thread counts differ significantly. The 265F offers 20 cores and 20 threads. The 358H provides 16 cores and 16 threads. Neither uses hyper-threading, so thread counts equal core counts. Clock speeds follow the expected pattern for their classes: the 265F has a 2.40 GHz base and 5.30 GHz boost, while the 358H runs at 1.90 GHz base and 4.80 GHz boost.

Cache hierarchies share the same per-core L1 (192 KB per core) and L2 (3 MB per core) allocations, but L3 differs. The 265F has 30 MB shared L3, while the 358H has 18 MB shared L3. The mobile chip makes up some ground in memory bandwidth: it supports LPDDR5X with 153.6 GB/s, versus DDR5 at 102.4 GB/s for the desktop part. Both use dual-channel memory buses and lack ECC support.

PCIe connectivity strongly favors the desktop chip. The 265F provides Gen 5 with 20 lanes (CPU only), while the 358H offers Gen 5 with just 4 lanes (CPU only). Integrated graphics also differ: the 265F has no integrated GPU (N/A), whereas the 358H includes Arc B390 graphics.

The 358H carries a 25 W TDP, reflecting its mobile positioning. The 265F has a 65 W TDP, typical for a mainstream desktop processor. Neither chip has an unlocked multiplier. The 265F was released on 2025-01-06 with a launch MSRP of $379. The 358H has a release date of 2026-01-04 and no recorded launch MSRP. The production status for both is listed as active.

The Verdict

The data directs users toward the Intel Core Ultra 7 265F for any compute-heavy workload. Its 20 cores, 5.30 GHz boost clock, 30 MB L3 cache, and 65 W desktop thermal headroom produce results that the 16-core, 4.80 GHz, 18 MB L3 mobile chip cannot approach. The 123.9% R23 multi-core lead and 184.9% single-core lead are decisive. The desktop part also sits at the 93rd percentile among all CPUs in the database, while the mobile chip ranks at the 87th percentile.

The 358H makes sense only in its native context: a 25 W mobile processor with integrated Arc B390 graphics and a 153.6 GB/s memory bus for LPDDR5X. For thin-and-light laptops where power efficiency matters more than absolute performance, it delivers competitive physics scores (3021 versus 3172, only 5% behind) and reasonable single-thread Passmark results (4124 versus 4750, 15.2% behind). But the benchmark data shows no scenario where the mobile chip wins outright.

Users who need desktop-grade multi-core rendering, data compression, encryption, or math throughput should select the 265F. Users who require a low-power mobile processor with integrated graphics should consider the 358H, accepting the substantial performance trade-offs documented above.

FAQ

Q: Which processor has more cores and threads?

A: The Intel Core Ultra 7 265F has 20 cores and 20 threads. The Intel Core Ultra X7 358H has 16 cores and 16 threads.

Q: How much faster is the 265F in Cinebench R23 multi-core?

A: The 265F scores 41980 versus 18747 for the 358H, a 123.9% advantage.

Q: Do both processors use the same manufacturing process?

A: Both use a 3 nm process, but the 265F is fabricated by TSMC while the 358H is fabricated by Intel.

Q: Which chip supports integrated graphics?

A: The Intel Core Ultra X7 358H includes Arc B390 integrated graphics. The Intel Core Ultra 7 265F has no integrated GPU.

Q: What is the memory bandwidth difference?

A: The 358H supports LPDDR5X with 153.6 GB/s bandwidth. The 265F supports DDR5 with 102.4 GB/s bandwidth.

Q: How close is the Passmark physics result between the two?

A: The 265F scores 3172 and the 358H scores 3021, a margin of just 5%.

Head-to-Head Benchmarks

The largest single-core margin appears in Cinebench R23 single-core, where the 265F scores 5926 against 2080 for the 358H. That 184.9% delta is the biggest win in the entire comparison. Cinebench R15 single-core follows with 597 versus 301.5, a 98% edge. These results highlight the clock speed and IPC advantages of the desktop Arrow Lake design.

Multi-core rendering shows similarly dramatic gaps. Cinebench R23 multi-core delivers 41980 versus 18747 (123.9% ahead). Cinebench R20 multi-core shows 17631 against 12011, a 46.8% lead. Cinebench R15 multi-core gives 4231 versus 3027, also 46.8% ahead.

Passmark math workloads favor the 265F heavily. Floating point math scores 173855 versus 103842, a 67.4% advantage. Integer math delivers 138078 versus 83147, 66.1% ahead. Extended instructions show 39235 versus 27274, a 43.9% edge.

Data processing tests follow the same pattern. Data compression scores 507018 versus 332508, a 52.5% lead. Random string sorting shows 62439 versus 40357, 54.7% ahead. Data encryption delivers 39468 versus 26046, a 51.5% advantage.

Multithreaded Passmark results put the 265F at 49410 versus 33802, a 46.2% lead. Prime number finding gives 416 versus 337, a 23.4% edge. Single-thread Passmark shows 4750 versus 4124, a 15.2% advantage. The narrowest result is physics, with 3172 versus 3021, just 5% apart.

The nearest rivals in the database provide context. The 265F averages 64438 in benchmark scores, sitting 0.3% behind the Intel Core Ultra 7 265 (64640), 0.6% ahead of the AMD EPYC 7343 (64202), 0.6% behind the AMD EPYC 4464P (64823), and 0.6% ahead of the Intel Core i9-13900KS (64051). The 358H averages 40967, placing it 0.6% behind the AMD Ryzen AI 5 PRO 440 (41208), 0.6% behind the Intel Core Ultra 7 356H (41215), 0.6% ahead of the AMD Ryzen AI 5 PRO 435G (40718), and 0.7% behind the Intel Core Ultra 7 366H (41263). These figures confirm that the 265F competes with high-end desktop and server parts, while the 358H sits in the mobile mid-range.