For now, Nvidia’s GeForce RTX 4090 graphics card comfortably sits on top of our GPU hierarchy, but as ever, the company is hard at work to offer something even more powerful — a seemingly inevitable GeForce RTX 4090 Ti graphics card. According to renowned hardware leaker @kopite7kimi, this product will use the AD102 graphics processor in a nearly full configuration, and it will consume up to 600W of power.
According to tweeted details, the Nvidia’s GeForce RTX 4090 Ti model PG136/139-SKU310 features AD102 GPU with 18,176 CUDA cores, a 96M L2cache and a 384-bit memory bus. Unlike the professional and datacenter-oriented RTX 6000 Ada Generation graphics card with 48GB of memory — which uses the AD102 in a similar configuration — the product aimed at gamers and prosumers will ‘only’ carry 24GB of GDDR6X memory, with a 24 GT/s data transfer speed, which is enough for all games, at least for now.
| GPU | FP32 CUDA Cores | Memory Configuration | TBP | MSRP | |
| GeForce RTX 4090 Ti | AD102 | 18176 (?) | 24GB 384-bit 24 GT/s GDDR6X (?) | 600W (?) | ? |
| GeForce RTX 4090 | AD102 | 16384 | 24GB 384-bit 21 GT/s GDDR6X | 450W | $1,599 |
| GeForce RTX 4080 | AD103 | 9728 | 16GB 256-bit 22.4 GT/s GDDR6X | 320W | $1,199 |
| GeForce RTX 4070 Ti | AD104 | 7680 | 12GB 192-bit 21 GT/s GDDR6X | 285W | $799 |
| GeForce RTX 4070 | AD104 | 5888 (?) | 12GB 192-bit 21 GT/s GDDR6X | 250W (?) | ? |
| GeForce RTX 4060 Ti | AD106 | 4352 (?) | 8GB 128-bit 18 GT/s GDDR6 | 160W (?) | <$500? |
| GeForce RTX 3070 | GA104 | 5888 | 8GB 256-bit 14 GT/s GDDR6 | 220W | $499 |
Power consumption of the new flagship graphics card from Nvidia will be up to 600W, according to the leaker, so expect it to feature an extremely sophisticated cooling system. In fact, it is logical to expect at least some of Nvidia’s board partners to ship their GeForce RTX 4090 Ti versions with hybrid closed-loop cooling systems.
Note that the AD102 GPU has 18,432 CUDA cores in total, so the GeForce RTX 4090 Ti will use almost all of them. At the moment, we are unsure whether Nvidia plans to ever use AD102 with all of its stream processors enabled, since it’s likely tough to find enough perfect GPUs with all cores working properly at high clock speeds.
It also remains to be seen how much Nvidia will charge for the GeForce RTX 4090 Ti. Keep in mind that the GeForce RTX 4090 is currently the world’s best graphics card for gaming, and that has no direct rivals. So it’s very likely that the GeForce RTX 4090 Ti will be more expensive than the current flagship (which often sells for more than $2,000). We can only wonder how much will the Titanium variant is set to cost. But unless you’re a professional user who can use a card like this in place of a much pricier option like the RTX 6000, it’s unlikely you’ll find the RTX 4090 Ti’s street price all that appealing.