AMD Instinct MI308X vs AMD Radeon RX 9060 Comparison

Head-to-Head Benchmarks

The database contains no direct head-to-head benchmark entries for the AMD Instinct MI308X and the AMD Radeon RX 9060. The Instinct MI308X has no recorded benchmark scores, no average benchmark score, and no nearest rivals listed. Its percentile versus all GPUs is 50, which indicates a median standing in the overall database distribution, but this ranking is not supported by any individual test scores. The Radeon RX 9060, in contrast, has a full set of recorded measurements across multiple suites. Its average benchmark score is 16014, and its percentile versus all GPUs is 59.

Because the Instinct MI308X lacks any benchmark results, the comparison must rely on architectural specifications and the recorded performance of the RX 9060 relative to its nearest rivals. The RX 9060's nearest rival is the NVIDIA GeForce RTX 3060 Ti, which holds an average score of 16129. The RX 9060 trails that card by 0.7 percent, a margin so small that the two effectively trade places depending on the workload. The RX 9060 leads the AMD Radeon R9 370X by 1 percent, with the R9 370X scoring 15862. It also leads the AMD Radeon RX 7700 by 1 percent, with the RX 7700 at 15852, and the AMD Radeon Pro W5500 by 2.1 percent, with the Pro W5500 at 15679.

The RX 9060's individual test scores show a specific strength profile. The highest score comes from Geekbench OpenCL at 88183, which reflects raw compute throughput. Geekbench Vulkan follows at 39476, indicating strong graphics API performance. The PassMark G3D score is 17631, and PassMark GPU compute is 9919. The DirectX-focused PassMark tests are lower, with DirectX 9 at 280, DirectX 11 at 182, DirectX 10 at 104, and DirectX 12 at 44. The PassMark G2D score is 1002. These results indicate that the card performs best in compute-oriented workloads and older DirectX titles, while its newer DirectX 12 score is comparatively modest.

The Instinct MI308X cannot be placed into these comparisons because no test scores exist in the database. Its percentile rank of 50 is the only performance-related figure recorded, and it carries no average benchmark score. The data therefore supports a quantitative analysis of the RX 9060 alone, while the MI308X can only be assessed through its specifications.

FAQ

Q: How does the AMD Radeon RX 9060 compare to its closest rival in the database?

A: The RX 9060 has an average benchmark score of 16014. Its nearest rival, the NVIDIA GeForce RTX 3060 Ti, averages 16129, so the RX 9060 trails by 0.7 percent. It leads the AMD Radeon R9 370X and the AMD Radeon RX 7700 by 1 percent each, and the AMD Radeon Pro W5500 by 2.1 percent.

Q: Does the AMD Instinct MI308X have any recorded benchmark scores?

A: No. The database lists no benchmarks for the MI308X, and its average benchmark score is 0. Its percentile versus all GPUs is 50, but this figure is not supported by any individual test results.

Q: What is the highest benchmark score recorded for the AMD Radeon RX 9060?

A: The highest score is 88183 in Geekbench OpenCL. The next highest is Geekbench Vulkan at 39476, followed by PassMark G3D at 17631 and PassMark GPU compute at 9919.

Q: What is the memory configuration of the AMD Instinct MI308X?

A: The MI308X uses 192 GB of HBM3 memory on an 8192-bit bus, with a bandwidth of 5.32 TB/s. The memory clock is 1300 MHz, which translates to 5.2 Gbps effective.

Q: What is the memory configuration of the AMD Radeon RX 9060?

A: The RX 9060 uses 8 GB of GDDR6 memory on a 128-bit bus, with a bandwidth of 288.0 GB/s. The memory clock is 2250 MHz, which translates to 18 Gbps effective.

Q: Which GPU has the higher FP32 throughput?

A: The AMD Instinct MI308X delivers 81.72 TFLOPS FP32, while the AMD Radeon RX 9060 delivers 21.43 TFLOPS FP32. The MI308X is substantially ahead in this metric.

Architecture Differences

The AMD Instinct MI308X and the AMD Radeon RX 9060 are built on different architectures and process nodes. The MI308X uses the CDNA 3.0 architecture with the Aqua Vanjaram chip, fabricated on a 5 nm process at TSMC. The RX 9060 uses the RDNA 4.0 architecture with the Navi 44 chip, fabricated on a 4 nm process at TSMC. The MI308X belongs to the Instinct (MIx) generation, while the RX 9060 belongs to the Navi IV (RX 9000) generation and the Radeon RX 9000 series.

The transistor counts differ sharply. The MI308X contains 153,000 million transistors on a die size of 1017 mm², giving a transistor density of 150.4 million per mm². The RX 9060 contains 29,700 million transistors on a die size of 199 mm², giving a transistor density of 149.2 million per mm². The two chips have nearly identical transistor densities, but the MI308X die is more than five times larger in area and holds over five times the transistor count.

The compute resources diverge significantly. The MI308X has 19456 shading units and 1216 texture mapping units, with no ROPs listed and no ray tracing cores. The RX 9060 has 1792 shading units, 112 texture mapping units, 64 ROPs, and 28 ray tracing cores. The MI308X reports a pixel rate of 0 MPixel/s, consistent with its lack of ROPs, while the RX 9060 reports 191.4 GPixel/s. The texture rates are 2,553.6 GTexel/s for the MI308X and 334.9 GTexel/s for the RX 9060.

The memory subsystems are also built for different purposes. The MI308X uses HBM3 with 192 GB capacity, an 8192-bit bus, and 5.32 TB/s bandwidth. The RX 9060 uses GDDR6 with 8 GB capacity, a 128-bit bus, and 288.0 GB/s bandwidth. The MI308X memory bandwidth is roughly 18 times that of the RX 9060. The MI308X memory clock is 1300 MHz (5.2 Gbps effective), while the RX 9060 memory clock is 2250 MHz (18 Gbps effective), but the RX 9060's narrow bus limits its total bandwidth.

Clock speeds follow a different pattern. The MI308X has a base clock of 1000 MHz and a boost clock of 2100 MHz. The RX 9060 has a base clock of 1700 MHz, a boost clock of 2990 MHz, and a game clock of 2400 MHz. The RX 9060 runs at higher frequencies across the board, but the MI308X compensates with far more compute units.

The MI308X has no display outputs and no API support listed for DirectX, OpenGL, or Vulkan. The RX 9060 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, and offers 1x HDMI 2.1b and 2x DisplayPort 2.1a outputs. The MI308X uses an OAM module slot width with no power connectors listed, while the RX 9060 is a dual-slot card with a single 8-pin power connector. The suggested PSU is 1150 W for the MI308X and 300 W for the RX 9060. The TDP is 750 W for the MI308X and 132 W for the RX 9060.

The Verdict

The data supports a clear division of roles. The AMD Instinct MI308X is a compute-oriented accelerator with 192 GB of HBM3 memory, 81.72 TFLOPS FP32, and 5.32 TB/s of memory bandwidth. Its 750 W TDP, OAM module form factor, and lack of display outputs indicate it is designed for server or data center installations where graphics output is unnecessary. The absence of any benchmark scores in the database means its real-world performance cannot be quantified here, but its specifications position it for high-capacity memory workloads and large-scale compute tasks.

The AMD Radeon RX 9060 is a consumer graphics card with 8 GB of GDDR6 memory, 21.43 TFLOPS FP32, and 288.0 GB/s bandwidth. It delivers display outputs, full API support, and a 132 W TDP. Its recorded benchmark scores place it near the NVIDIA GeForce RTX 3060 Ti, trailing by 0.7 percent, and ahead of the AMD Radeon R9 370X, RX 7700, and Pro W5500 by 1 to 2.1 percent. The RX 9060 is the only one of the two with measurable performance data in the database.

The MI308X has a percentile rank of 50 versus all GPUs, which is lower than the RX 9060's 59. However, this percentile is not supported by any benchmark scores for the MI308X, so it should not be read as a direct performance comparison. The RX 9060's percentile of 59 is backed by 10 recorded tests across multiple suites.

Specification Differences

The two GPUs differ in nearly every recorded specification. The MI308X uses the CDNA 3.0 architecture, while the RX 9060 uses RDNA 4.0. The process nodes are 5 nm and 4 nm, respectively, both from TSMC. The MI308X has 153,000 million transistors on a 1017 mm² die, while the RX 9060 has 29,700 million transistors on a 199 mm² die. Transistor density is nearly identical: 150.4M per mm² for the MI308X versus 149.2M per mm² for the RX 9060.

Clock speeds differ. The MI308X runs at 1000 MHz base and 2100 MHz boost. The RX 9060 runs at 1700 MHz base, 2990 MHz boost, and 2400 MHz game clock. Memory clocks are 1300 MHz (5.2 Gbps effective) for the MI308X and 2250 MHz (18 Gbps effective) for the RX 9060.

Memory configurations diverge completely. The MI308X has 192 GB of HBM3 on an 8192-bit bus with 5.32 TB/s bandwidth. The RX 9060 has 8 GB of GDDR6 on a 128-bit bus with 288.0 GB/s bandwidth. The MI308X has 19456 shading units and 1216 TMUs, with no ROPs. The RX 9060 has 1792 shading units, 112 TMUs, 64 ROPs, and 28 ray tracing cores. Pixel rates are 0 MPixel/s for the MI308X and 191.4 GPixel/s for the RX 9060. Texture rates are 2,553.6 GTexel/s and 334.9 GTexel/s, respectively.

FP32 and FP16 throughput are both 81.72 TFLOPS for the MI308X and 21.43 TFLOPS for the RX 9060, with both offering a 1:1 ratio. TDP is 750 W for the MI308X and 132 W for the RX 9060. The MI308X uses an OAM Module slot width with no power connectors, while the RX 9060 is dual-slot with a single 8-pin connector. Suggested PSU is 1150 W for the MI308X and 300 W for the RX 9060. Both use PCIe 5.0 x16. The MI308X has no display outputs, while the RX 9060 has 1x HDMI 2.1b and 2x DisplayPort 2.1a. The MI308X lists no API support, while the RX 9060 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Release dates also differ. The MI308X was released on 2023-12-05, and the RX 9060 on 2025-08-04. The MI308X predecessor is listed as Radeon Instinct, while the RX 9060 predecessor is Navi III. The RX 9060 production status is Active; the MI308X production status is not recorded.

Where Each One Wins

The AMD Instinct MI308X wins in memory capacity, memory bandwidth, and raw compute throughput. Its 192 GB of HBM3 memory and 5.32 TB/s bandwidth are orders of magnitude beyond the RX 9060's 8 GB and 288.0 GB/s. Its FP32 throughput of 81.72 TFLOPS is nearly four times the RX 9060's 21.43 TFLOPS. Its texture rate of 2,553.6 GTexel/s is more than seven times the RX 9060's 334.9 GTexel/s. These specifications suit workloads that need to hold very large datasets or move data at extreme speeds, such as large-scale compute tasks with no display requirement.

The AMD Radeon RX 9060 wins in every area related to consumer graphics use. It has display outputs, while the MI308X has none. It supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, while the MI308X lists no API support. It has 64 ROPs and 28 ray tracing cores, which the MI308X lacks entirely. Its pixel rate of 191.4 GPixel/s is functional, whereas the MI308X records 0 MPixel/s. Its boost clock of 2990 MHz is higher than the MI308X's 2100 MHz. Its TDP of 132 W is far below the MI308X's 750 W, and its suggested PSU of 300 W is far below the MI308X's 1150 W. The RX 9060 is the only one of the two with recorded benchmark scores, and those scores place it within 1 percent of the NVIDIA GeForce RTX 3060 Ti.

The RX 9060 also wins in portability and integration. It is a dual-slot card with a single 8-pin power connector, while the MI308X is an OAM module with no power connectors listed. The RX 9060 has a production status of Active, while the MI308X production status is not recorded. The RX 9060 has a higher percentile rank versus all GPUs at 59, compared to the MI308X's 50, though the MI308X percentile is not supported by benchmark data.