NVIDIA N1X 40SM vs Lisuan Tech LX PRO Comparison

Where Each One Wins

The recorded data splits these two accelerators into completely different usage profiles. The NVIDIA N1X 40SM is an integrated graphics processor (IGP) built on the Blackwell 2.0 architecture, designed to ride the PCIe 5.0 x16 bus with a massive 128 GB of LPDDR5X memory. Its strengths lie in capacity and texture throughput. The Lisuan Tech LX PRO, by contrast, is a dual-slot, 225 W discrete card using the TrueGPU architecture, with 24 GB of GDDR6 on a 192-bit bus, and it wins outright in pixel throughput, memory bandwidth, and raw FP16 compute.

The N1X 40SM takes the win in texture rate, posting 750.7 GTexel/s against the LX PRO's 384.0 GTexel/s. That is roughly a 1.96x advantage in texel fill, which favors workloads that hammer texture sampling, such as certain content-creation pipelines or scientific visualization that relies on dense texture fetches. The N1X also fields 320 texture mapping units (TMUs) versus 192 on the LX PRO, and its shading unit count of 5120 is lower than the LX PRO's 6144, but the TMU disparity is what drives the texture rate gap.

The LX PRO wins decisively in pixel rate, delivering 192.0 GPixel/s compared to the N1X's 93.84 GPixel/s. That is a 2.05x advantage in raster output, and it pairs with 96 ROPs on the LX PRO against just 40 on the N1X. For any workload that resolves pixels, such as high-resolution rendering or display output to multiple monitors, the LX PRO has the clear edge. Its memory bandwidth also leads: 432.0 GB/s versus 273.2 GB/s, a 1.58x margin, which matters for large data sets that exceed the N1X's narrower 256-bit bus.

FP32 compute is nearly a tie. The N1X posts 24.02 TFLOPS, and the LX PRO posts 24.58 TFLOPS, a 2.3% lead for the LX PRO. That is within noise territory for most real workloads. FP16 is where the LX PRO separates itself: it delivers 49.15 TFLOPS with a 2:1 ratio, while the N1X delivers 24.02 TFLOPS at a 1:1 ratio. That means the LX PRO has a 2.05x advantage in half-precision math, which is directly relevant to AI inference and training loops that use FP16 accumulations.

The N1X compensates with a full 128 GB of memory, which is 5.33x the LX PRO's 24 GB. That capacity is the N1X's primary differentiator. For workloads that require holding enormous models or data sets in local memory, the N1X is the only option between these two. The LX PRO counters with GDDR6 speed and a smaller footprint, but it cannot match the N1X's sheer memory pool.

Architecture Differences

The two chips come from fundamentally different design lineages. The N1X uses the GB20B chip, built on Blackwell 2.0, at TSMC's 5 nm process, with a die size of 382 mm². The LX PRO uses the 7G105 chip, built on TrueGPU, at TSMC's 6 nm process, with a die size that is not recorded. Both use TSMC as the foundry, but the process nodes differ by one generation step.

The N1X is an integrated part, marked as "IGP" in slot width, with no power connectors and no recorded TDP. It uses PCIe 5.0 x16 and has a single HDMI display output. The LX PRO is a discrete dual-slot card that is 248 mm long, 118 mm high, and 48 mm wide, with a 225 W TDP, a single 16-pin power connector, a suggested 550 W power supply, and four DisplayPort 1.4a outputs. The N1X has no recorded API support (DirectX, OpenGL, Vulkan all listed as N/A), while the LX PRO supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.3.

Core counts differ in every category. The N1X has 5120 shading units, 320 TMUs, 40 ROPs, 40 RT cores, and 160 tensor cores. The LX PRO has 6144 shading units, 192 TMUs, 96 ROPs, and no recorded RT or tensor core counts. The N1X's tensor core count of 160 is unique to that chip, but the LX PRO does not list any tensor core hardware. The LX PRO's shading unit advantage is 1.2x, while the N1X's TMU advantage is 1.67x and its ROP deficit is 2.4x.

Clock behavior also diverges. The N1X has a base clock of 741 MHz and a boost of 2346 MHz, with memory at 1067 MHz (8.5 Gbps effective). The LX PRO has no recorded base or boost clocks for the core, but its memory runs at 2250 MHz (18 Gbps effective), which is more than double the N1X's effective memory speed. The N1X's memory is LPDDR5X on a 256-bit bus, while the LX PRO uses GDDR6 on a 192-bit bus.

Release dates are close: the N1X launched on 2026-05-31, and the LX PRO launched on 2026-03-16. Both are listed as Active in production status. The N1X is part of the Blackwell IGP (N1x) generation, while the LX PRO belongs to the 7G100 generation. Neither has recorded predecessor or successor parts.

Head-to-Head Benchmarks

The database has no recorded benchmark scores for either part, and no head-to-head benchmark entries exist. Both parts sit at the 50th percentile against all GPUs, with an average benchmark score of 0. That means the comparison must be made entirely from specification-derived rates: pixel rate, texture rate, FP32, FP16, and memory bandwidth.

The largest single advantage belongs to the LX PRO in pixel rate. At 192.0 GPixel/s versus 93.84 GPixel/s, the LX PRO is 2.05x faster. This is driven by 96 ROPs versus 40, a 2.4x ROP advantage, and the LX PRO's higher core clock potential, though the core clock is not recorded. For any rendering that resolves to a framebuffer, the LX PRO delivers double the output rate.

The N1X wins texture rate by a similar margin in the opposite direction. At 750.7 GTexel/s versus 384.0 GTexel/s, the N1X is 1.95x faster. This comes from 320 TMUs versus 192, and the N1X's boost clock of 2346 MHz, which likely sustains high texture fetch throughput. The LX PRO's texture rate is limited by its lower TMU count despite having more shading units.

Memory bandwidth favors the LX PRO: 432.0 GB/s versus 273.2 GB/s, a 1.58x lead. The LX PRO's GDDR6 at 18 Gbps effective on a 192-bit bus delivers more bandwidth than the N1X's LPDDR5X at 8.5 Gbps effective on a 256-bit bus. The N1X's wider bus cannot compensate for the LX PRO's faster memory clock.

FP16 compute is the LX PRO's second-largest win. At 49.15 TFLOPS versus 24.02 TFLOPS, the LX PRO is 2.05x faster, thanks to its 2:1 FP16 ratio. The N1X runs FP16 at a 1:1 ratio, meaning it offers no half-precision throughput advantage over FP32. For AI workloads that use FP16, the LX PRO has the clear edge.

FP32 is effectively a tie: 24.58 TFLOPS for the LX PRO versus 24.02 TFLOPS for the N1X, a 2.3% difference. Both parts deliver roughly the same single-precision throughput, so any FP32-bound workload will see near-identical performance.

Memory capacity is the N1X's biggest win: 128 GB versus 24 GB, a 5.33x advantage. The N1X can hold more than five times the data locally, which matters for large language models or scientific data sets that exceed 24 GB. The LX PRO cannot approach this capacity.

The Verdict

The data points to two distinct purchasing decisions based on workload type. The Lisuan Tech LX PRO is the stronger choice for any task that depends on pixel throughput, memory bandwidth, or half-precision compute. Its 2.05x pixel rate advantage, 1.58x bandwidth lead, and 2.05x FP16 throughput make it the clear pick for real-time rendering, high-resolution output, and FP16-heavy AI inference.

The NVIDIA N1X 40SM is the choice for texture-bound workloads and massive memory capacity. Its 1.95x texture rate advantage and 5.33x memory capacity are unmatched by the LX PRO. For workloads that stream large data sets into a single pool of 128 GB, such as certain simulation or data analytics tasks, the N1X is the only viable option between these two.

The LX PRO also has the advantage of a full API stack: DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.3. The N1X lists no API support at all. That means software compatibility is a decisive factor for any application that relies on those graphics APIs. The N1X's single HDMI output versus the LX PRO's four DisplayPort 1.4a outputs further reinforces the LX PRO as the display-oriented part.

Power and form factor also separate them. The LX PRO is a discrete 225 W dual-slot card with a 16-pin connector and a 550 W suggested power supply. The N1X is an IGP with no power connectors and no recorded TDP, meaning it draws power from the motherboard and requires no additional cooling beyond the system design. The N1X is the lower-integration-cost option, while the LX PRO demands a proper power supply and chassis space.

Neither part has benchmark scores in the database, so the verdict rests on specification-derived rates and architectural features. The LX PRO leads in four of six comparative metrics: pixel rate, FP32, FP16, and memory bandwidth. The N1X leads in two: texture rate and memory capacity. The LX PRO's wins are larger in aggregate, but the N1X's capacity advantage is so large that it may dominate for specific use cases.

FAQ

Q: Which part has higher memory bandwidth?

A: The Lisuan Tech LX PRO leads with 432.0 GB/s, compared to the NVIDIA N1X 40SM's 273.2 GB/s, a 1.58x margin.

Q: How much memory does each part have?

A: The NVIDIA N1X 40SM has 128 GB of LPDDR5X, while the Lisuan Tech LX PRO has 24 GB of GDDR6. The N1X has 5.33x the capacity.

Q: Is the N1X faster in texture processing?

A: Yes. The N1X posts 750.7 GTexel/s, which is 1.95x the LX PRO's 384.0 GTexel/s, driven by 320 TMUs versus 192.

Q: Which part supports DirectX and Vulkan?

A: Only the Lisuan Tech LX PRO, which supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.3. The NVIDIA N1X 40SM lists no API support.

Q: What is the FP16 performance difference?

A: The LX PRO delivers 49.15 TFLOPS at a 2:1 ratio, while the N1X delivers 24.02 TFLOPS at a 1:1 ratio. The LX PRO is 2.05x faster in half-precision.

Q: Are the parts similar in FP32 compute?

A: Nearly identical. The LX PRO posts 24.58 TFLOPS, and the N1X posts 24.02 TFLOPS, a 2.3% difference.

Specification Differences

The following fields differ between the two parts:

• Architecture: NVIDIA N1X uses Blackwell 2.0; Lisuan Tech LX PRO uses TrueGPU.
• Chip: NVIDIA N1X uses GB20B; LX PRO uses 7G105.
• Process Node: NVIDIA N1X is 5 nm; LX PRO is 6 nm.
• Die Size: NVIDIA N1X is 382 mm²; LX PRO is unknown.
• Base Clock: NVIDIA N1X is 741 MHz; LX PRO has no recorded base clock.
• Boost Clock: NVIDIA N1X is 2346 MHz; LX PRO has no recorded boost clock.
• Memory Clock: NVIDIA N1X is 1067 MHz (8.5 Gbps effective); LX PRO is 2250 MHz (18 Gbps effective).
• Memory Size: NVIDIA N1X is 128 GB; LX PRO is 24 GB.
• Memory Type: NVIDIA N1X is LPDDR5X; LX PRO is GDDR6.
• Memory Bus Width: NVIDIA N1X is 256 bit; LX PRO is 192 bit.
• Memory Bandwidth: NVIDIA N1X is 273.2 GB/s; LX PRO is 432.0 GB/s.
• Shading Units: NVIDIA N1X has 5120; LX PRO has 6144.
• TMUs: NVIDIA N1X has 320; LX PRO has 192.
• ROPs: NVIDIA N1X has 40; LX PRO has 96.
• RT Cores: NVIDIA N1X has 40; LX PRO has no recorded RT cores.
• Tensor Cores: NVIDIA N1X has 160; LX PRO has no recorded tensor cores.
• Pixel Rate: NVIDIA N1X is 93.84 GPixel/s; LX PRO is 192.0 GPixel/s.
• Texture Rate: NVIDIA N1X is 750.7 GTexel/s; LX PRO is 384.0 GTexel/s.
• FP32: NVIDIA N1X is 24.02 TFLOPS; LX PRO is 24.58 TFLOPS.
• FP16: NVIDIA N1X is 24.02 TFLOPS (1:1); LX PRO is 49.15 TFLOPS (2:1).
• TDP: NVIDIA N1X is unknown; LX PRO is 225 W.
• Slot Width: NVIDIA N1X is IGP; LX PRO is dual-slot.
• Power Connectors: NVIDIA N1X has none; LX PRO has 1x 16-pin.
• Suggested PSU: NVIDIA N1X has none recorded; LX PRO is 550 W.
• Bus Interface: NVIDIA N1X is PCIe 5.0 x16; LX PRO is PCIe 4.0 x16.
• Display Outputs: NVIDIA N1X has 1x HDMI; LX PRO has 4x DisplayPort 1.4a.
• APIs: NVIDIA N1X has no DirectX, OpenGL, or Vulkan support; LX PRO has DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.3.
• Dimensions: NVIDIA N1X has no recorded dimensions; LX PRO is 248 mm long, 118 mm high, 48 mm wide.
• Release Date: NVIDIA N1X is 2026-05-31; LX PRO is 2026-03-16.
• Generation: NVIDIA N1X is Blackwell IGP (N1x); LX PRO is 7G100.