NVIDIA RTX 5000 Ada Generation Laptop GPU

The NVIDIA RTX 5000 Ada Generation Laptop GPU is a professional graphics card for notebooks. It is based on the consumer RTX 4090 Laptop GPU with 9,728 CUDA cores, 76 ray tracing cores and 304 4th generation tensor cores. The RTX 5000 also uses 16 GB of memory with a 256-bit bus, but the memory supports ECC error correction (optional, reduces the size to 15360MB). The clock rates are always slightly lower for the professional RTX models, which is why the performance is also behind the RTX 4090 Mobile. However, as with the RTX 4090, this depends on the TDP provided by the notebook. According to Nvidia, the RTX 5000 can be configured from 80 to 175 watts with correspondingly large performance differences.

Despite the similar name, the GPU is not comparable to the older NVIDIA Quadro RTX 5000 (Turing).

The AD103 chip used is manufactured by TSMC in the 4N process (5nm).

RTX Ada Generation Series NVIDIA RTX 5000 Ada Generation Laptop GPU 9728 @ 0.93 - 1.68 GHz 256 Bit @ 20000 MHz NVIDIA RTX 4000 Ada Generation Laptop GPU 7424 192 Bit @ 16000 MHz NVIDIA RTX 3500 Ada Generation Laptop GPU 5120 192 Bit @ 16000 MHz NVIDIA RTX 3000 Ada Generation Laptop GPU 4608 128 Bit @ 16000 MHz NVIDIA RTX 2000 Ada Generation Laptop GPU 3072 128 Bit @ 16000 MHz
Architecture	Ada Lovelace
Pipelines	9728 - unified
TMUs	304
ROPs	112
Raytracing Cores	76
Tensor / AI Cores	304
Core Speed	930 - 1680 (Boost) MHz
Cache	L2: 64 MB
Memory Speed	20000 effective = 2250 MHz
Memory Bus Width	256 Bit
Memory Type	GDDR6
Max. Amount of Memory	16 GB
Shared Memory	no
Memory Bandwidth	576 GB/s
API	DirectX 12 Ultimate, Shader 6.7, OpenGL 4.6, OpenCL 3.0, Vulkan 1.3
Power Consumption	150 Watt (80 - 150 Watt TGP)
Transistor Count	76.3 Billion
technology	4 nm
PCIe	4 x16/x8
Displays	4 Displays (max.), HDMI 2.1, DisplayPort 1.4a
Notebook Size	large
Date of Announcement	21.03.2023
Link to Manufacturer Page	nvdam.widen.net
Predecessor	RTX A5500 Laptop GPU

Benchmarks