NVIDIA Reveals Complete Specifications for GeForce RTX 5070 Ti and RTX 5070 GPUs Featuring GB203 and GB205 “Blackwell” Chips

NVIDIA has officially announced the complete specifications for the GeForce RTX 5070 Ti and RTX 5070 graphics cards, part of the new Blackwell series. This revelation fills in the gaps left after the initial disclosure at CES, where pricing and release dates were also hinted at.

NVIDIA GeForce RTX 5070 Ti & RTX 5070: Key Specifications Revealed

Following NVIDIA’s unveiling of the RTX 50 series GPUs, we now have an official PDF detailing the full specs for the RTX 5070 Ti and RTX 5070 models. The GeForce RTX 5070 Ti is positioned as the third most powerful GPU in the Blackwell family, succeeding the RTX 4070 Ti. It features the GB203 GPU die, which is also utilized in the RTX 5080, leading to some similarities in specifications such as the 16 GB of GDDR7 VRAM. However, the RTX 5070 Ti operates at a speed of 28 Gbps, slightly lower than the 30 Gbps found in the RTX 5080.

Full specifications of NVIDIA GeForce RTX 5070 Ti & RTX 5070 GPU

Conversely, the NVIDIA GeForce RTX 5070 is developed on the GB205 die, making it the least powerful GPU unveiled. This model will replace the RTX 4070, although it is notable that it features a downgraded GPU die compared to the previous Ampere and Ada generation’s 104 die. The RTX 5070 does, however, offer improvements in areas like the CUDA cores, Tensor cores, and RT cores.

While the VRAM capacity remains unchanged at 12 GB with a 192-bit bus, it now benefits from a faster GDDR7 memory. The boost clock for the RTX 5070 rises to 2512 MHz—higher than the RTX 4070’s 2475 MHz. Conversely, the RTX 5070 Ti sees a decrease in boost clock speed at 2452 MHz, compared to the RTX 4070 Ti’s 2610 MHz. Noteworthy enhancements are observed in memory configuration, where the RTX 5070 Ti features 16 GB of GDDR7 memory versus 12 GB of GDDR6X and a generous 256-bit memory bus.

NVIDIA GeForce RTX 5070 Ti specifications — Image Source: NVIDIA

Power ratings stand at 300W for the RTX 5070 Ti and 250W for the non-Ti model. Custom AIB (add-in-board) editions may require more power due to enhanced cooling solutions and clock adjustments. NVIDIA anticipates releasing the RTX 5070 Ti on February 20, followed by the RTX 5070 several weeks later in March. Both models will come with a $50 price reduction compared to predecessors, with special variants potentially costing more.

NVIDIA GeForce RTX 5070 Ti: Enhanced Graphics Experience

The NVIDIA GeForce RTX 5070 Ti is equipped with a trimmed-down GB203 GPU, featuring 8960 CUDA cores. Its memory setup includes the same 16 GB capacity as the RTX 4070 Ti SUPER, but with newer GDDR7 chips that deliver a 78% increase in bandwidth over the 4070 Ti, achieving up to 896 GB/s. The boost clock is finalized at 2.45 GHz.

Performance-wise, the RTX 5070 Ti promises a substantial twofold improvement over the RTX 4070 Ti, priced at $749, which offers a competitive edge against the 4070 Ti and 4070 Ti SUPER due to its enhancements and support for DLSS 4 features.

NVIDIA GeForce RTX 5070: Competitive Entry-Level Graphics Card

Meanwhile, the NVIDIA GeForce RTX 5070 will utilize the GB205 GPU attached to the PG147 reference PCB. This variant operates with up to 6144 CUDA cores and maintains a 192-bit memory bus, paralleling the existing 12 GB RTX 4070.

NVIDIA GeForce RTX 5070 specifications — Image Source: NVIDIA

The RTX 5070 achieves a memory speed of 28 Gbps, promoting up to 672 GB/s of bandwidth—a 33% enhancement over the RTX 4070 and 4070 SUPER GPUs. It operates with a total board power of 250W, reflecting a 14% increase compared to the RTX 4070 SUPER’s 220W.

Notably, the NVIDIA GeForce RTX 5070 is reported to outperform the RTX 4090 while maintaining a more attractive price point of $549. This positions it as a compelling option within the high-end market, especially as it competes against AMD’s newly launched RX 9070 series. Both the RTX 5070 Ti and RTX 5070 are classified as potent choices for 1440p gaming, whereas the RTX 5080 targets 4K capabilities.

As a reference point, the peak configurations of the Blackwell “RTX 50″GPUs are as follows:

GB202 (FULL) -> 192 SMs / 24576 Cores (100%) / RTX 5090 -> 170 SMs / 21760 Cores (88.5%)
GB203 (FULL) -> 84 SMs / 10752 Cores (100%) / RTX 5080 -> 84 SMs / 10752 Cores (100%)
GB205 (FULL) -> 50 SMs / 6400 Cores (100%) / RTX 5070 -> TBD SMs / TBD Cores (TBD %)

Preparations are in place for the launch of the NVIDIA GeForce RTX 5080 on January 30, followed by the RTX 5070 Ti and RTX 5070 in February. All models will be available in custom editions alongside the popular Founders Edition variants.

NVIDIA RTX 50 Founders Edition Models:

Overview of NVIDIA GeForce RTX 50 GPU Specifications (Preliminary):

Graphics Card Name	NVIDIA GeForce RTX 5090	NVIDIA GeForce RTX 5080	NVIDIA GeForce RTX 5070 Ti	NVIDIA GeForce RTX 5070	NVIDIA GeForce RTX 5060 Ti	NVIDIA GeForce RTX 5060
GPU Name	Blackwell GB202-300	Blackwell GB203-400	Blackwell GB203-300-A1	Blackwell GB205-300-A1	Blackwell GB206?	Blackwell GB206?
GPU SMs	170 (192 Full)	84 (84 Full)	70 (84 Full)	50 (50 SM Full)	TBD	TBD
GPU Cores	21760 (+33%)	10752 (+11%)	8960 (+16%)	6144	TBD	TBD
Clock Speeds	2.41 GHz	2.62 GHz	2.45 GHz	2.51 GHz	TBD	TBD
L2 Cache	TBD	TBD	TBD	TBD	TBD	TBD
Memory Capacity	32 GB GDDR7 (+33%)	16 GB GDDR7 (0%)	16 GB GDDR7	12 GB GDDR7	16 GB GDDR7	8 GB GDDR7
Memory Bus	512-bit (+33%)	256-bit (0%)	256-bit (0%)	192-bit (0%)	128-bit (0%)	128-bit (0%)
Memory Speed	28 Gbps	30 Gbps	28 Gbps	28 Gbps	28 Gbps?	28 Gbps?
Bandwidth	1792 GB/s	960GB/s	896 GB/s	672 GB/s	448 GB/s	448 GB/s
TBP	575W (+27%)	360W (+12.5%)	300W (+10.5%)	250W (+14%)	TBD	TBD
Power Interface	1 12V-2×6 (16-Pin)	1 12V-2×6 (16-Pin)	1 12V-2×6 (16-Pin)	1 12VHPWR (16-Pin)	1 12VHPWR (16-Pin)	1 12VHPWR (16-Pin)
Launch	January 2025	January 2025	February 2025	February 2025	March-April 2025	March-April 2025
Price	$1999 US	$999 US	$749 US	$549 US	TBD	TBD