Nvidia GeForce RTX 5070 Laptop

The Nvidia GeForce RTX 5070 Laptop is a mobile graphics card from the Blackwell family for laptops, which was unveiled at CES 2025. It is based on the GB206 chip with 4,608 shader / CUDA cores clocked at 1425 - 2347 MHz (depending on the model). The graphics memory used is 8 GB GDDR7 with a 128-bit memory bus.

Architecture

The Nvidia GeForce RTX 5070 laptop has new Tensor (5th generation) and RT cores (4th generation) with a performance of 798 AI TOPS. The Blackwell architecture is the first with the new DLSS 4 Multi Frame Generation, which can generate more frames per second thanks to improved AI capabilities, while using less memory and delivering better image quality. Nvidia has also managed to improve the performance and output quality of the DLSS, Frame Generation and Ray Reconstruction features (which were also available on the previous RTX 4000 series). In addition, the new models now support Reflex 2 mode with low latency and Frame Warp.

Performance

The gaming performance of the RTX 5070 Mobile is heavily dependent on the cooling and TDP settings of the laptop. As a result, the maximum clock rate can also vary from 1425 to 2347 MHz. On average, the performance is usually slightly above the old RTX 4070 Laptop GPU but below the RTX 3080 Ti laptop. A new feature of the Blackwell architecture is the support for multi-frame generation (MFG) in DLSS4. This allows significantly higher frame rates to be achieved, but without MFG you should already reach over 50 fps to achieve a playable latency.

The chip is manufactured in the current TSMC 4N process.

GeForce RTX 5000 Laptop GPU Series GeForce RTX 5090 Laptop 10496 @ 0.99 - 1.52 GHz 256 Bit @ 28000 MHz GeForce RTX 5080 Laptop 8192 @ 0.98 - 2.29 GHz 256 Bit @ 28000 MHz GeForce RTX 5070 Ti Laptop 5888 @ 0.85 - 2.22 GHz 192 Bit @ 32000 MHz GeForce RTX 5070 Laptop 4608 @ 0.91 - 2.35 GHz 128 Bit @ 32000 MHz GeForce RTX 5060 Laptop 3328 @ 0.95 - 1.46 GHz 128 Bit @ 24000 MHz GeForce RTX 5050 Laptop 2560 @ 1.02 - 2.66 GHz 128 Bit @ 14000 MHz
Codename	GN22-X6
Architecture	Blackwell
Pipelines	4608 - unified
TMUs	144
ROPs	48
Raytracing Cores	36
Tensor / AI Cores	144
Core Speed	907 - 2347 (Boost) MHz
Theoretical Performance	23.22 TFLOPS FP16, 23.22 TFLOPS FP32
AI GPU	798 TOPS INT8
Cache	L1: 4.608 KB, L2: 32.768 KB
Memory Speed	32000 effective = 2000 MHz
Memory Bus Width	128 Bit
Memory Type	GDDR7
Max. Amount of Memory	8 GB
Shared Memory	no
Memory Bandwidth	512 GB/s
API	DirectX 12_2, OpenGL 4.6, OpenCL 3.0, Vulkan 1.4
Power Consumption	100 Watt (50 - 100 Watt TGP)
technology	5 nm
PCIe	5
Displays	4 Displays (max.), HDMI 2.1b, DisplayPort 2.1b
Notebook Size	medium sized
Date of Announcement	02.01.2025
Predecessor	NVIDIA GeForce RTX 4070 Laptop GPU

Benchmarks