AMD Instinct MI325X vs AMD Radeon RX 9060 XT LP Comparison

AMD
RADEON

AMD Instinct MI325X

CORE STATE Aqua Vanjaram
VRAM 256 GB
CLOCK SPEED 2100 MHz
TDP 1000 W
BUS WIDTH 8192 bit
ARCHITECTURE CDNA 3.0
nm
PROCESS 5 nm
LAUNCH DATE 2024
VS
AMD
RADEON

Radeon RX 9060 XT LP

CORE STATE Navi 44
VRAM 16 GB
CLOCK SPEED 3050 MHz
TDP 140 W
BUS WIDTH 128 bit
ARCHITECTURE RDNA 4.0
nm
PROCESS 4 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

geekbench_opencl
N/A
88,183
geekbench_vulkan
N/A
39,476

Analysis: AMD Instinct MI325X vs AMD Radeon RX 9060 XT LP

AMD Instinct MI325X is a compute-oriented accelerator built on CDNA 3.0, while AMD Radeon RX 9060 XT LP is a graphics-focused card on RDNA 4.0. The benchmark database records no overlapping test scores for the two, as the MI325X has no benchmark entries and the RX 9060 XT LP has results in Geekbench OpenCL and Vulkan. The MI325X targets massive parallel compute workloads with 256 GB of HBM3e memory and 81.72 TFLOPS FP32, whereas the RX 9060 XT LP delivers 24.99 TFLOPS FP32 with 16 GB of GDDR6 and full graphics API support including DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4. The data shows a clear functional split: one is an accelerator without display outputs, the other is a dual-slot graphics card with display connectivity.

Where Each One Wins

The AMD Instinct MI325X wins in raw compute throughput and memory capacity. Its FP32 performance of 81.72 TFLOPS is over three times the 24.99 TFLOPS of the RX 9060 XT LP, and its FP16 performance matches FP32 at 81.72 TFLOPS (1:1), indicating strong compute density for AI and scientific workloads. The MI325X packs 256 GB of HBM3e memory with an 8192-bit bus, delivering 6.14 TB/s of bandwidth, compared to the RX 9060 XT LP's 16 GB of GDDR6 on a 128-bit bus at 322.3 GB/s. Texture rate also favors the MI325X at 2,553.6 GTexel/s versus 390.4 GTexel/s. The MI325X has 19,456 shading units and 1,216 TMUs, dwarfing the RX 9060 XT LP's 2,048 shading units and 128 TMUs.

The AMD Radeon RX 9060 XT LP wins in graphics rendering, power efficiency, and physical footprint. It has 64 ROPs and a pixel rate of 195.2 GPixel/s, while the MI325X has 0 ROPs and a pixel rate of 0 MPixel/s, meaning the MI325X cannot rasterize images at all. The RX 9060 XT LP includes 32 ray tracing cores, which the MI325X lacks entirely. The RX 9060 XT LP supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, while the MI325X lists N/A for all three APIs. The RX 9060 XT LP has a 140 W TDP and requires a 300 W suggested PSU, whereas the MI325X has a 1000 W TDP and a 1400 W suggested PSU. The RX 9060 XT LP is a dual-slot card with a single 8-pin power connector and display outputs (1x HDMI 2.1b, 2x DisplayPort 2.1a), while the MI325X is an OAM module with no power connectors and no display outputs.

The Verdict

The data directs each product to a distinct audience. The AMD Instinct MI325X is for server and data center environments where massive memory pools and extreme FP32/FP16 throughput matter more than any graphics capability. Its 256 GB HBM3e and 6.14 TB/s bandwidth are the decisive factors for large model inference or scientific simulation. The RX 9060 XT LP is for desktop or workstation builds requiring a compact dual-slot card with modern graphics APIs and ray tracing. Its 89th percentile ranking among all GPUs in the database, with an average benchmark score of 63,830, places it near NVIDIA CMP 30HX (63,842, delta 0%) and AMD Radeon RX 7600M (63,775, delta 0.1%), and slightly ahead of AMD Radeon Pro WX 9100 (64,212, delta -0.6%). The MI325X has a 50th percentile ranking with an average benchmark score of 0 due to no recorded tests, so direct performance comparison is impossible.

For compute-only tasks, the MI325X is the clear choice based on raw specifications. For any task involving rendering, gaming, or display output, the RX 9060 XT LP is the only viable option. The MI325X has no display outputs and no graphics API support, making it unsuitable for interactive workloads. The RX 9060 XT LP, with its 16 GB memory and 24.99 TFLOPS, handles graphics workloads efficiently within a 140 W envelope. The production status of the MI325X is not recorded, while the RX 9060 XT LP is marked as active in production.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark results between the two cards, as the MI325X has no benchmark entries and the RX 9060 XT LP has only Geekbench OpenCL and Vulkan scores. The RX 9060 XT LP scores 88,183 in Geekbench OpenCL and 39,476 in Geekbench Vulkan. These scores place it in the 89th percentile among all GPUs, with an average score of 63,830. Its nearest rivals include NVIDIA CMP 30HX (average score 63,842, delta 0%), AMD Radeon RX 7600M (63,775, delta 0.1%), AMD Radeon Pro Vega 56 (63,693, delta 0.2%), and AMD Radeon Pro WX 9100 (64,212, delta -0.6%). The delta percentages show the RX 9060 XT LP is effectively tied with these cards, staying within 0.6% of all four.

The absence of MI325X benchmarks means no recorded wins for either side in head-to-head tests. The wins column shows 0 for both items. Specification comparisons provide the only measurable differences. The MI325X leads in shading units (19,456 vs 2,048), TMUs (1,216 vs 128), texture rate (2,553.6 GTexel/s vs 390.4 GTexel/s), FP32 (81.72 TFLOPS vs 24.99 TFLOPS), FP16 (81.72 TFLOPS vs 24.99 TFLOPS), memory size (256 GB vs 16 GB), memory bandwidth (6.14 TB/s vs 322.3 GB/s), and memory bus width (8192 bit vs 128 bit). The RX 9060 XT LP leads in ROPs (64 vs 0), pixel rate (195.2 GPixel/s vs 0 MPixel/s), ray tracing cores (32 vs 0), base clock (1380 MHz vs 1000 MHz), boost clock (3050 MHz vs 2100 MHz), game clock (2450 MHz vs none), memory clock (2518 MHz 20.1 Gbps effective vs 1500 MHz 6 Gbps effective), and process node (4 nm vs 5 nm).

FAQ

Q: Which card has higher FP32 compute performance?

A: The AMD Instinct MI325X delivers 81.72 TFLOPS FP32, which is 3.27 times the 24.99 TFLOPS of the AMD Radeon RX 9060 XT LP.

Q: Can the AMD Instinct MI325X output video to a display?

A: No. The MI325X has no display outputs, while the RX 9060 XT LP provides 1x HDMI 2.1b and 2x DisplayPort 2.1a.

Q: How does the RX 9060 XT LP compare to its nearest rivals in the database?

A: The RX 9060 XT LP has an average benchmark score of 63,830, placing it in the 89th percentile. It sits within 0.6% of NVIDIA CMP 30HX (63,842), AMD Radeon RX 7600M (63,775), AMD Radeon Pro Vega 56 (63,693), and AMD Radeon Pro WX 9100 (64,212).

Q: What memory types and capacities do the two cards use?

A: The MI325X uses 256 GB of HBM3e with an 8192-bit bus, while the RX 9060 XT LP uses 16 GB of GDDR6 on a 128-bit bus.

Q: Which card supports ray tracing?

A: Only the RX 9060 XT LP supports ray tracing, with 32 ray tracing cores. The MI325X has no ray tracing cores listed.

Q: What is the power consumption difference?

A: The MI325X has a 1000 W TDP with a 1400 W suggested PSU, while the RX 9060 XT LP has a 140 W TDP with a 300 W suggested PSU.

Architecture Differences

The MI325X uses the Aqua Vanjaram chip with CDNA 3.0 architecture, while the RX 9060 XT LP uses Navi 44 with RDNA 4.0 architecture. Both are manufactured by TSMC, but on different nodes: the MI325X at 5 nm and the RX 9060 XT LP at 4 nm. Transistor counts differ dramatically: the MI325X has 153,000 million transistors on a 1017 mm² die, while the RX 9060 XT LP has 29,700 million transistors on a 199 mm² die. Transistor density is similar at 150.4M per mm² for the MI325X and 149.2M per mm² for the RX 9060 XT LP.

The MI325X belongs to the Instinct (MIx) generation with a predecessor of Radeon Instinct, and its architecture targets compute acceleration with no graphics pipeline. The RX 9060 XT LP belongs to the Radeon RX 9000 series, Navi IV generation, with a predecessor of Navi III. The MI325X has no ROPs, no ray tracing cores, and no graphics API support, confirming its compute-only design. The RX 9060 XT LP includes 64 ROPs, 32 ray tracing cores, and full API support for DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Clock behavior also reflects design goals. The MI325X has a base clock of 1000 MHz and boost of 2100 MHz, with no game clock. The RX 9060 XT LP has a base clock of 1380 MHz, boost of 3050 MHz, and a game clock of 2450 MHz, showing it is tuned for sustained graphics workloads. Memory clocks follow the same pattern: the MI325X runs at 1500 MHz (6 Gbps effective) for HBM3e, while the RX 9060 XT LP runs at 2518 MHz (20.1 Gbps effective) for GDDR6.

Specification Differences

The two cards differ in nearly every measured specification. The MI325X has 19,456 shading units, 1,216 TMUs, and 0 ROPs, while the RX 9060 XT LP has 2,048 shading units, 128 TMUs, and 64 ROPs. The MI325X has no ray tracing cores; the RX 9060 XT LP has 32. Pixel rate is 0 MPixel/s for the MI325X versus 195.2 GPixel/s for the RX 9060 XT LP. Texture rate is 2,553.6 GTexel/s for the MI325X versus 390.4 GTexel/s for the RX 9060 XT LP.

Memory configuration is the largest divide. The MI325X has 256 GB of HBM3e on an 8192-bit bus with 6.14 TB/s bandwidth. The RX 9060 XT LP has 16 GB of GDDR6 on a 128-bit bus with 322.3 GB/s bandwidth. Power requirements differ by an order of magnitude: 1000 W TDP for the MI325X versus 140 W TDP for the RX 9060 XT LP. Suggested PSU ratings are 1400 W and 300 W respectively. The MI325X is an OAM module with no power connectors, while the RX 9060 XT LP is a dual-slot card with one 8-pin connector.

Physical and interface differences matter for deployment. The MI325X has no display outputs; the RX 9060 XT LP has 1x HDMI 2.1b and 2x DisplayPort 2.1a. Both use PCIe 5.0 x16. The MI325X has a release date of 2024-10-09, while the RX 9060 XT LP has a release date of 2025-12-16. The RX 9060 XT LP has an active production status; the MI325X does not have a recorded production status. No launch MSRP is recorded for either product. The MI325X has no benchmark entries in the database, while the RX 9060 XT LP has Geekbench OpenCL (88,183) and Vulkan (39,476) scores, leading to an average benchmark score of 63,830 and an 89th percentile ranking.

DETAILED SPECIFICATIONS

SPECIFICATION
Instinct MI325X
RX 9060 XT LP
Core Specs
Shading Units
19,456
2,048 -89.5%
Shaders
19,456
2,048 -89.5%
TMUs
1,216
128 -89.5%
ROPs
0
64 +∞%
Compute Units
304
32 -89.5%
Clocks
Base Clock
1000 MHz
1380 MHz
Boost Clock
2100 MHz
3050 MHz
Game Clock
—
2450 MHz
Memory Clock
1500 MHz 6 Gbps effective
2518 MHz 20.1 Gbps effective
Memory
Memory Size
256 GB
16 GB
VRAM (MB)
262,144
16,384 -93.8%
Memory Type
HBM3e
GDDR6
Memory Bus
8192 bit
128 bit
Bandwidth
6.14 TB/s
322.3 GB/s
Cache
L1 Cache
16 KB (per CU)
—
L2 Cache
16 MB
4 MB
L3 Cache
256 MB
32 MB
L0 Cache
—
32 KB per WGP
Performance
Pixel Rate
0 MPixel/s
195.2 GPixel/s
Texture Rate
2,553.6 GTexel/s
390.4 GTexel/s
FP32 (TFLOPS)
81.72 TFLOPS
24.99 TFLOPS
FP64 (TFLOPS)
40.86 TFLOPS (1:2)
780.8 GFLOPS (1:32)
FP16 (TFLOPS)
81.72 TFLOPS (1:1)
24.99 TFLOPS (1:1)
AI/RT
RT Cores
—
32
Matrix Cores
1,216
64 -94.7%
Power
TDP
1000 W
140 W
TDP (W)
1,000
140 -86.0%
Suggested PSU
1400 W
300 W
Power Connectors
None
1x 8-pin
Architecture
Architecture
CDNA 3.0
RDNA 4.0
GPU Name
Aqua Vanjaram
Navi 44
Generation
Instinct (MIx)
Navi IV (RX 9000)
Process Size
5 nm
4 nm
Transistors
153,000 million
29,700 million
Die Size
1017 mm²
199 mm²
Foundry
TSMC
TSMC
Density
150.4M / mm²
149.2M / mm²
AMD MCM
MCM
2
—
API Support
DirectX
—
12 Ultimate (12_2)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
2.2
Shader Model
—
6.9
Physical
Slot Width
OAM Module
Dual-slot
Outputs
No outputs
1x HDMI 2.1b2x DisplayPort 2.1a
Bus Interface
PCIe 5.0 x16
PCIe 5.0 x16
Other
Production
—
Active
Predecessor
Radeon Instinct
Navi III
View Instinct MI325X Details View Radeon RX 9060 XT LP Details