AMD Radeon RX 6750 GRE 12 GB vs NVIDIA GeForce RTX 5090 SE Comparison

AMD
RADEON

AMD Radeon RX 6750 GRE 12 GB

CORE STATE Navi 22
VRAM 12 GB
CLOCK SPEED 2581 MHz
TDP 250 W
BUS WIDTH 192 bit
ARCHITECTURE RDNA 2.0
nm
PROCESS 7 nm
LAUNCH DATE 2023
VS
NVIDIA
GEFORCE

GeForce RTX 5090 SE

CORE STATE GB202
VRAM 24 GB
CLOCK SPEED 2377 MHz
TDP 500 W
BUS WIDTH 384 bit
ARCHITECTURE Blackwell 2.0
nm
PROCESS 5 nm
LAUNCH DATE 2026

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12
2,402
N/A
geekbench_vulkan
108,994
N/A

Analysis: AMD Radeon RX 6750 GRE 12 GB vs NVIDIA GeForce RTX 5090 SE

Head-to-Head Benchmarks

The database contains two benchmark entries for the AMD Radeon RX 6750 GRE 12 GB: a 3DMark Steel Nomad DX12 score of 2402 and a Geekbench Vulkan score of 108994. The NVIDIA GeForce RTX 5090 SE currently has no recorded benchmark scores in the database, with an average benchmark score of zero and no nearest rivals listed. This absence of data means direct numerical comparisons between these two cards cannot be established from the recorded measurements.

The AMD Radeon RX 6750 GRE 12 GB carries an average benchmark score of 55698 across all recorded tests. Its percentile ranking places it at the 87th percentile among all GPUs tracked in the database. The nearest rivals provide context for this score. The AMD Radeon 8060S sits 0.1% below with an average score of 55757, representing a nearly identical performance level. The NVIDIA GeForce RTX 5080 trails by 0.7% with an average score of 56083. The AMD Radeon Pro W5700X comes in 1.6% higher with an average score of 54828, while the NVIDIA GeForce RTX 4080 sits 2.7% higher with an average score of 54247. These delta values indicate that the RX 6750 GRE 12 GB operates within a tight band of performance relative to these four comparable products, with all deltas within a range of roughly three percentage points.

For the RTX 5090 SE, the database records a percentile rank of 50 against all GPUs, but this figure accompanies an average benchmark score of zero. The percentile value appears inconsistent with the zero score, yet the database provides no additional benchmark entries to resolve this discrepancy. Consequently, the head-to-head comparison relies on architectural and specification analysis rather than measured performance data.

The RX 6750 GRE 12 GB delivers 13.21 TFLOPS of FP32 compute and 26.43 TFLOPS of FP16 compute at a 2:1 ratio. The RTX 5090 SE delivers 66.94 TFLOPS for both FP32 and FP16 at a 1:1 ratio. This represents a substantial compute advantage for the NVIDIA card, with FP32 output approximately five times higher. Pixel fill rates follow a similar pattern: the RX 6750 GRE produces 165.2 GPixel/s while the RTX 5090 SE produces 380.3 GPixel/s, more than double. Texture fill rates show 413.0 GTexel/s for the AMD card versus 1,045.9 GTexel/s for the NVIDIA card, a ratio of roughly 2.5 to one.

Memory bandwidth differences are equally pronounced. The RX 6750 GRE uses 12 GB of GDDR6 memory across a 192-bit bus, yielding 432.0 GB/s of bandwidth. The RTX 5090 SE uses 24 GB of GDDR7 memory across a 384-bit bus, yielding 1.34 TB/s of bandwidth. The NVIDIA card offers double the capacity and more than three times the bandwidth. Clock speeds tell a different story. The AMD card runs a base clock of 2321 MHz and a boost clock of 2581 MHz, with a game clock of 2439 MHz. The NVIDIA card runs a base clock of 1740 MHz and a boost clock of 2377 MHz. The AMD card holds the advantage in both base and boost frequencies, though the architectural differences between RDNA 2.0 and Blackwell 2.0 make direct clock comparisons less meaningful.

The Verdict

The recorded data shows that the RX 6750 GRE 12 GB occupies a firmly established performance tier, sitting at the 87th percentile with an average score of 55698. Its nearest rivals cluster within a few percentage points, indicating consistent, competitive performance against contemporary products in the same class. The RTX 5090 SE, by contrast, has no average score, no rivals, and a percentile rank of 50 that does not align with any measured performance data. From the database alone, the RTX 5090 SE cannot be positioned relative to any other GPU.

Users seeking a card with verified benchmark results and a clear competitive standing should consider the RX 6750 GRE 12 GB. The data confirms it matches or slightly exceeds several high-end alternatives. The 87th percentile ranking places it among the top tier of all GPUs tracked. The RTX 5090 SE offers no such verification in the database, making any performance claim unsupported by measurements.

Users prioritizing raw specifications on paper would find the RTX 5090 SE delivers dramatically higher compute, memory bandwidth, and memory capacity. The FP32 throughput of 66.94 TFLOPS, the 1.34 TB/s memory bandwidth, and the 24 GB of memory all far exceed the corresponding figures on the AMD card. The RTX 5090 SE also carries 440 tensor cores and 110 RT cores, compared to 40 RT cores and no tensor cores on the RX 6750 GRE. Specifications alone suggest the NVIDIA card targets workloads requiring massive parallel compute and large memory footprints. The RX 6750 GRE suits users who value verified performance data and a known competitive position within the database.

The release dates separate the two products by more than two years. The RX 6750 GRE launched on October 16, 2023, while the RTX 5090 SE launched on December 31, 2025. The AMD card remains active in production, and the NVIDIA card also shows active production status. Both cards use dual-slot designs, with identical lengths of 267 mm and widths of 40 mm. The RX 6750 GRE stands 110 mm tall, while the RTX 5090 SE stands 111 mm tall.

Architecture Differences

The RX 6750 GRE uses the Navi 22 chip built on RDNA 2.0 architecture, fabricated on a 7 nm process at TSMC with 17,200 million transistors on a 335 mm² die. The transistor density reaches 51.3 million transistors per square millimeter. The RTX 5090 SE uses the GB202 chip built on Blackwell 2.0 architecture, fabricated on a 5 nm process at TSMC with 92,200 million transistors on a 750 mm² die. The transistor density reaches 122.9 million transistors per square millimeter. The NVIDIA chip contains more than five times the transistor count and more than twice the die area, with a density advantage of roughly 2.4 times.

The RX 6750 GRE belongs to the Radeon RX 6000 series and the Navi II generation, with its predecessor listed as Navi and successor as Navi III. The RTX 5090 SE belongs to the GeForce 50-series and the GeForce 50 generation, with its predecessor listed as GeForce 40 and successor as GeForce 60. Both architectures support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The hardware implementations differ in ray tracing and tensor capabilities. The RX 6750 GRE includes 40 ray tracing cores and no tensor cores. The RTX 5090 SE includes 110 ray tracing cores and 440 tensor cores. The tensor cores on the NVIDIA card enable AI acceleration features that the AMD card cannot match at the hardware level.

Compute unit organization differs significantly. The RX 6750 GRE contains 2560 shading units, 160 texture mapping units, and 64 render output units. The RTX 5090 SE contains 14080 shading units, 440 texture mapping units, and 160 render output units. The NVIDIA card provides 5.5 times the shading units, 2.75 times the texture mapping units, and 2.5 times the render output units. These ratios align with the FP32 and texture rate differences observed in the specification data.

Memory architecture reflects the generational gap. The RX 6750 GRE uses GDDR6 memory at 2250 MHz with 18 Gbps effective data rate, delivering 432.0 GB/s over a 192-bit bus. The RTX 5090 SE uses GDDR7 memory at 1750 MHz with 28 Gbps effective data rate, delivering 1.34 TB/s over a 384-bit bus. The newer GDDR7 standard and wider bus provide the NVIDIA card with a memory bandwidth advantage that exceeds its compute advantage relative to the AMD card.

Specification Differences

The power requirements separate these cards clearly. The RX 6750 GRE carries a TDP of 250 W, uses one 6-pin and one 8-pin power connector, and recommends a 600 W power supply. The RTX 5090 SE carries a TDP of 500 W, uses a single 16-pin power connector, and recommends a 900 W power supply. The NVIDIA card consumes twice the power and requires a substantially larger power supply.

The bus interfaces differ by generation. The RX 6750 GRE uses PCIe 4.0 x16, while the RTX 5090 SE uses PCIe 5.0 x16. Display outputs also differ: the AMD card provides one HDMI 2.1 and three DisplayPort 1.4a outputs, while the NVIDIA card provides one HDMI 2.1b and three DisplayPort 2.1b outputs. Both cards measure 267 mm in length and 40 mm in width, with the AMD card at 110 mm height and the NVIDIA card at 111 mm height.

Memory configurations show the largest gap. The RX 6750 GRE offers 12 GB of GDDR6, while the RTX 5090 SE offers 24 GB of GDDR7. The launch MSRP for the RX 6750 GRE is 334 USD, and the launch MSRP for the RTX 5090 SE is 1,499 USD. The production status for both cards is listed as active.

The FP16 compute ratio highlights a key architectural difference. The RX 6750 GRE achieves 26.43 TFLOPS of FP16 at a 2:1 ratio, meaning it processes FP16 at half the rate of a native 1:1 implementation. The RTX 5090 SE achieves 66.94 TFLOPS of FP16 at a 1:1 ratio, meaning it processes FP16 at the same rate as FP32. This difference matters for workloads that rely heavily on half-precision arithmetic.

FAQ

Q: What is the average benchmark score for each card?

A: The AMD Radeon RX 6750 GRE 12 GB has an average benchmark score of 55698. The NVIDIA GeForce RTX 5090 SE has an average benchmark score of 0.

Q: How does the RX 6750 GRE compare to its nearest rivals?

A: The RX 6750 GRE sits 0.1% below the AMD Radeon 8060S, 0.7% below the NVIDIA GeForce RTX 5080, 1.6% above the AMD Radeon Pro W5700X, and 2.7% above the NVIDIA GeForce RTX 4080.

Q: What are the memory specifications for both cards?

A: The RX 6750 GRE uses 12 GB of GDDR6 memory on a 192-bit bus with 432.0 GB/s bandwidth. The RTX 5090 SE uses 24 GB of GDDR7 memory on a 384-bit bus with 1.34 TB/s bandwidth.

Q: Which card has higher clock speeds?

A: The RX 6750 GRE has a base clock of 2321 MHz and a boost clock of 2581 MHz. The RTX 5090 SE has a base clock of 1740 MHz and a boost clock of 2377 MHz. The AMD card operates at higher frequencies in both cases.

Q: What is the FP32 compute performance difference?

A: The RX 6750 GRE delivers 13.21 TFLOPS of FP32 compute, while the RTX 5090 SE delivers 66.94 TFLOPS. The NVIDIA card provides roughly five times the FP32 throughput.

Q: What power supply is recommended for each card?

A: The RX 6750 GRE has a TDP of 250 W and recommends a 600 W power supply. The RTX 5090 SE has a TDP of 500 W and recommends a 900 W power supply.

Q: How many ray tracing cores does each card have?

A: The RX 6750 GRE includes 40 ray tracing cores. The RTX 5090 SE includes 110 ray tracing cores. The NVIDIA card also includes 440 tensor cores, which the AMD card does not have.

DETAILED SPECIFICATIONS

SPECIFICATION
RX 6750 GRE 12 GB
RTX 5090 SE
Core Specs
Shading Units
2,560
14,080 +450.0%
Shaders
2,560
14,080 +450.0%
TMUs
160
440 +175.0%
ROPs
64
160 +150.0%
Compute Units
40
—
SM Count
—
110
Clocks
Base Clock
2321 MHz
1740 MHz
Boost Clock
2581 MHz
2377 MHz
Game Clock
2439 MHz
—
Memory Clock
2250 MHz 18 Gbps effective
1750 MHz 28 Gbps effective
Memory
Memory Size
12 GB
24 GB
VRAM (MB)
12,288
24,576 +100.0%
Memory Type
GDDR6
GDDR7
Memory Bus
192 bit
384 bit
Bandwidth
432.0 GB/s
1.34 TB/s
Cache
L1 Cache
128 KB per Array
128 KB (per SM)
L2 Cache
3 MB
96 MB
L3 Cache
96 MB
—
L0 Cache
32 KB per WGP
—
Performance
Pixel Rate
165.2 GPixel/s
380.3 GPixel/s
Texture Rate
413.0 GTexel/s
1,045.9 GTexel/s
FP32 (TFLOPS)
13.21 TFLOPS
66.94 TFLOPS
FP64 (TFLOPS)
825.9 GFLOPS (1:16)
1,045.9 GFLOPS (1:64)
FP16 (TFLOPS)
26.43 TFLOPS (2:1)
66.94 TFLOPS (1:1)
AI/RT
RT Cores
40
110 +175.0%
Tensor Cores
—
440
Power
TDP
250 W
500 W
TDP (W)
250
500 +100.0%
Suggested PSU
600 W
900 W
Power Connectors
1x 6-pin + 1x 8-pin
1x 16-pin
Architecture
Architecture
RDNA 2.0
Blackwell 2.0
GPU Name
Navi 22
GB202
Generation
Navi II (RX 6000)
GeForce 50
Process Size
7 nm
5 nm
Transistors
17,200 million
92,200 million
Die Size
335 mm²
750 mm²
Foundry
TSMC
TSMC
Density
51.3M / mm²
122.9M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
2.1
3.0
CUDA
—
12.0
Shader Model
6.8
6.9
Physical
Slot Width
Dual-slot
Dual-slot
Length
267 mm 10.5 inches
267 mm 10.5 inches
Height
110 mm 4.3 inches
111 mm 4.4 inches
Outputs
1x HDMI 2.13x DisplayPort 1.4a
1x HDMI 2.1b3x DisplayPort 2.1b
Bus Interface
PCIe 4.0 x16
PCIe 5.0 x16
Other
Launch Price
334 USD
1,499 USD
Production
Active
Active
Predecessor
Navi
GeForce 40
Successor
Navi III
GeForce 60
View Radeon RX 6750 GRE 12 GB Details View GeForce RTX 5090 SE Details