AMD’s next-gen Radeon RX 7900 XT flagship GPU will leverage a bunch of advanced process and packaging technologies, including TSMC 5nm and 6nm nodes. Featuring the Navi 31 core, it will pack a total of seven dies. The primary Graphics Compute Die (GCD) will be based on the 5nm process with up to 12,288 stream processors across 96 CUs and 48 WGPs. The remaining six dies will be the MCDs (Memory Complex Dies) fabbed on the N6 node. Each of these will consist of a 64-bit memory controller paired with 32MB of L3 “Infinity” Cache.
GPU Name | RX 7700 XT (Navi 33) | RX 7800 XT (Navi 32) | RX 7900 XT (Navi 31) | RX 7950 XT (Navi 30?) |
---|---|---|---|---|
Process Node | 6nm | 5nm + 6nm | 5nm + 6nm | 5nm + 6nm |
Die Config | 6nm x1 | 5nm x1, 6nm x4 | 5nm x1, 6nm x5 | 5nm x1, 6nm x6 |
Shader Engines | 2 | 4 | 6 | 6 |
WGPs | 16 | 32 | 48 | 48 |
CUs | 32 | 64 | 96 | 96 |
Cores | 4,096 | 8,192 | 10,752? | 12,288 |
Memory Bus | 128-bit | 256-bit | 320-bit? | 384-bit? |
Memory Type | GDDR6 | GDDR6 | GDDR6 | GDDR6 |
Memory Capacity | 8GB | 16GB | 20GB? | 24GB? |
Infinity Cache | 64MB | 128MB | 192MB | 192MB? |
TGP | 200W | 250W? | 300? | 350W? |
Launch | Q4 2022 | Q1 2023 | Q4 2022 | 2023 |
For the fully enabled Navi 31 GPU, this results in a total L3 cache of 192MB and a bus width of 384-bit along with 16-32GB of GDDR6 18Gbps memory. According to @Kepler_L2 (Twitter), there will be certain models with even more Infinity Cache. These cache dies will be 3D stacked on the MCDs, each contributing 32MB for an overall SRAM of 384MB.

This should confer the Radeon RX 7900 XT with incredible memory bandwidth, much higher than the RTX 4090, internally. The latter should still be faster on account of the higher ALU count but I reckon the delta won’t be significant, even at higher resolutions.
Source: