AMD Ryzen AI Z2 Extreme GPU vs Lisuan Tech LX PRO Comparison

Where Each One Wins

The recorded data splits the two products cleanly along workload profiles. The AMD Ryzen AI Z2 Extreme GPU is a 28 W integrated-class part built for efficiency, while the Lisuan Tech LX PRO is a 225 W dual-slot add-in card built for throughput. The AMD side wins in power-sensitive configurations, offering a 16 GB LPDDR5X frame buffer on a 256 bit bus with 256.0 GB/s bandwidth. The Lisuan side wins in raw compute and memory capacity, carrying 24 GB GDDR6 on a 192 bit bus with 432.0 GB/s bandwidth.

In shading throughput, the Lisuan LX PRO delivers 24.58 TFLOPS FP32, which is 4.4 times the AMD part's 5.530 TFLOPS. The gap widens further in FP16, where the Lisuan reaches 49.15 TFLOPS with a 2:1 ratio, versus the AMD part's 5.530 TFLOPS at 1:1. Texture fill follows the same pattern: 384.0 GTexel/s for Lisuan versus 172.8 GTexel/s for AMD, a 2.2 times advantage. Pixel throughput favors Lisuan at 192.0 GPixel/s versus 129.6 GPixel/s, a 1.5 times margin.

The AMD part wins in connector and power simplicity. It uses no external power connectors and outputs video through a single USB Type-C port. The Lisuan card requires one 16-pin connector, a 550 W suggested PSU, and takes a dual-slot footprint of 248 mm length, 118 mm height, and 48 mm width. The AMD solution has no listed dimensions, indicating a mobile or embedded form factor.

Benchmark wins are evenly split at zero each in the head-to-head data, which reflects the absence of recorded benchmark scores rather than performance parity. The percentile rank against all GPUs is 50 for both, placing them at the median of the database's tracked products. The use-case split therefore comes from architecture and specifications: the AMD part targets devices where power draw and physical footprint matter, the Lisuan card targets workloads demanding high sustained throughput and large memory pools.

Architecture Differences

The two GPUs come from different design lineages. AMD uses the Strix Point chip with RDNA 3.5 architecture, built on TSMC's 4 nm process. The chip integrates 34,000 million transistors on a 233 mm² die, yielding a density of 145.9M transistors per mm². Lisuan uses the 7G105 chip with an architecture labeled TrueGPU, built on TSMC's 6 nm process. Transistor count and die size are not recorded for the Lisuan part, so density cannot be compared.

The AMD GPU has 1024 shading units, 64 texture mapping units, and 48 raster operation units. It also includes 16 dedicated ray tracing cores, a feature the Lisuan part does not list. The Lisuan GPU instead fields 6144 shading units, 192 TMUs, and 96 ROPs, with no ray tracing cores recorded. The shading unit count gives Lisuan a 6 to 1 advantage in raw ALU count, but the AMD architecture compensates with a higher process density and lower power envelope.

Memory architecture diverges sharply. AMD uses 16 GB of LPDDR5X at 1000 MHz with 8 Gbps effective transfer, over a 256 bit bus, producing 256.0 GB/s. Lisuan uses 24 GB of GDDR6 at 2250 MHz with 18 Gbps effective transfer, over a 192 bit bus, producing 432.0 GB/s. The Lisuan card has 50% more capacity and 68.75% more bandwidth, despite a narrower bus, because the GDDR6 memory clock runs much higher.

API support shows a slight difference in Vulkan versions. Both support DirectX 12 Ultimate (12_2) and OpenGL 4.6. AMD lists Vulkan 1.4, while Lisuan lists Vulkan 1.3. The AMD part also carries a 1:1 FP16 to FP32 ratio, meaning it does not gain extra throughput in half-precision work. The Lisuan part doubles FP16 throughput with a 2:1 ratio, making it more suited to workloads that can use half precision.

Power delivery and cooling differ completely. AMD operates at 28 W with no power connectors, suggesting a soldered or low-profile solution. Lisuan operates at 225 W with a single 16-pin connector and a 550 W suggested PSU. The Lisuan card is dual-slot with physical dimensions of 248 mm by 118 mm by 48 mm, while the AMD part has no recorded slot width or dimensions.

Release timing places the AMD part on 2025-10-15 and the Lisuan on 2026-03-16. Both are listed as Active production. Neither has a launch MSRP in the database.

FAQ

Q: Which GPU has more memory bandwidth?

A: The Lisuan Tech LX PRO has 432.0 GB/s bandwidth from 24 GB GDDR6 on a 192 bit bus. The AMD Ryzen AI Z2 Extreme GPU has 256.0 GB/s from 16 GB LPDDR5X on a 256 bit bus.

Q: Does the AMD part support ray tracing?

A: Yes, the AMD Ryzen AI Z2 Extreme GPU includes 16 ray tracing cores. The Lisuan Tech LX PRO has no ray tracing cores listed in the database.

Q: What is the FP32 compute difference between the two?

A: The Lisuan Tech LX PRO delivers 24.58 TFLOPS FP32, which is 4.4 times the AMD Ryzen AI Z2 Extreme GPU's 5.530 TFLOPS.

Q: Which GPU has a newer Vulkan version?

A: The AMD Ryzen AI Z2 Extreme GPU supports Vulkan 1.4, while the Lisuan Tech LX PRO supports Vulkan 1.3. Both support DirectX 12 Ultimate (12_2) and OpenGL 4.6.

Q: What power connector does the Lisuan card require?

A: The Lisuan Tech LX PRO uses one 16-pin power connector and has a suggested PSU of 550 W. The AMD part uses no external power connectors.

Q: What is the production status of both GPUs?

A: Both the AMD Ryzen AI Z2 Extreme GPU and the Lisuan Tech LX PRO are listed as Active production. The AMD part released on 2025-10-15, and the Lisuan released on 2026-03-16.

Specification Differences

The two GPUs differ across nearly every recorded specification except for their percentile rank, which is 50 for both.

Process node: AMD uses 4 nm, Lisuan uses 6 nm. Both are fabricated by TSMC.

Transistors: AMD has 34,000 million on a 233 mm² die. Lisuan has unknown transistor count and unknown die size.

Shading units: AMD has 1024, Lisuan has 6144. TMUs: AMD has 64, Lisuan has 192. ROPs: AMD has 48, Lisuan has 96. Ray tracing cores: AMD has 16, Lisuan has none recorded.

Clocks: AMD has a base clock of 800 MHz and a boost of 2700 MHz. Lisuan has no base or boost clock recorded. Memory clock: AMD runs at 1000 MHz with 8 Gbps effective, Lisuan runs at 2250 MHz with 18 Gbps effective.

Memory: AMD has 16 GB LPDDR5X on a 256 bit bus with 256.0 GB/s. Lisuan has 24 GB GDDR6 on a 192 bit bus with 432.0 GB/s.

Rates: AMD pixel rate is 129.6 GPixel/s, Lisuan is 192.0 GPixel/s. AMD texture rate is 172.8 GTexel/s, Lisuan is 384.0 GTexel/s. AMD FP32 is 5.530 TFLOPS, Lisuan is 24.58 TFLOPS. AMD FP16 is 5.530 TFLOPS at 1:1, Lisuan is 49.15 TFLOPS at 2:1.

Power and cooling: AMD TDP is 28 W with no power connectors and no slot width. Lisuan TDP is 225 W with one 16-pin connector, a 550 W suggested PSU, and dual-slot dimensions of 248 mm length, 118 mm height, and 48 mm width.

Bus interface: AMD has none recorded, Lisuan uses PCIe 4.0 x16.

Display outputs: AMD has one USB Type-C, Lisuan has four DisplayPort 1.4a.

APIs: Both support DirectX 12 Ultimate (12_2) and OpenGL 4.6. AMD supports Vulkan 1.4, Lisuan supports Vulkan 1.3.

Release dates: AMD on 2025-10-15, Lisuan on 2026-03-16.

Head-to-Head Benchmarks

The database records no head-to-head benchmark scores for these two GPUs, and neither has entries in the benchmark arrays. The wins count is zero for both sides. This means the comparison must rely on the specification-derived rates, which provide clear numerical deltas.

The largest single advantage for the Lisuan part appears in FP16 compute. At 49.15 TFLOPS, it exceeds the AMD part's 5.530 TFLOPS by a factor of 8.9. That is the widest margin in any measured capability. FP32 follows at 24.58 TFLOPS versus 5.530 TFLOPS, a 4.4 times gap. Texture rate gives Lisuan a 2.2 times lead at 384.0 GTexel/s versus 172.8 GTexel/s. Pixel rate gives a 1.5 times lead at 192.0 GPixel/s versus 129.6 GPixel/s.

Memory bandwidth favors Lisuan by a factor of 1.69, with 432.0 GB/s versus 256.0 GB/s. Capacity favors Lisuan by a factor of 1.5, with 24 GB versus 16 GB. The AMD part maintains a narrower bus width at 256 bit versus 192 bit, but the Lisuan memory clock of 2250 MHz with 18 Gbps effective more than compensates.

The AMD part does not win any recorded rate comparisons. Its advantages are qualitative: a smaller process node at 4 nm versus 6 nm, a lower TDP at 28 W versus 225 W, no power connector requirement, and a newer Vulkan version at 1.4 versus 1.3. It also has ray tracing cores, which the Lisuan lacks, though no benchmark data quantifies the impact.

Both GPUs sit at the 50th percentile against all GPUs in the database. With an average benchmark score of zero for each and no nearest rivals listed, the percentile value indicates a median position but does not reflect any measured performance event. The data suggests these are placeholder or early-entry records awaiting benchmark results.

The Verdict

The data points to a clear split by use case. The AMD Ryzen AI Z2 Extreme GPU is the choice for systems where power draw and physical integration matter. It operates at 28 W, requires no external power connector, uses a single USB Type-C display output, and has no recorded dimensions, indicating a compact or embedded installation. Its RDNA 3.5 architecture on a 4 nm process with 16 ray tracing cores makes it suitable for DirectX 12 Ultimate workloads in a low-power envelope.

The Lisuan Tech LX PRO is the choice for throughput-oriented tasks. It delivers 24.58 TFLOPS FP32, 49.15 TFLOPS FP16, 384.0 GTexel/s texture fill, and 192.0 GPixel/s pixel fill. Its 24 GB GDDR6 frame buffer with 432.0 GB/s bandwidth provides headroom for large datasets and high-resolution textures. The 225 W TDP and 550 W suggested PSU indicate a desktop card that requires a proper power supply and a dual-slot case.

For ray tracing, the AMD part has a structural advantage with 16 dedicated RT cores, while the Lisuan lists none. For half-precision compute, the Lisuan part has a structural advantage with a 2:1 FP16 ratio, doubling its throughput. Neither part has benchmark scores to confirm real-world performance, so the verdict rests on the recorded specifications and rates.

Users who need a self-contained, low-power GPU with ray tracing support should look at the AMD Ryzen AI Z2 Extreme GPU. Users who need maximum compute throughput, memory capacity, and bandwidth in a desktop system should look at the Lisuan Tech LX PRO. The percentile ranks are identical at 50, which the database records as a median position, but the specification deltas are large enough to separate the two clearly.