NovaPodRTX 4090
Flagship ConsumerNVIDIA · 24GB VRAM
Market
from $0.1343/hr
median $0.4940/hr·p10 $0.3876/hr
419 offers available
30 days
+10.53%
- Min
- $0.3400
- Avg
- $0.3730
- Max
- $0.4200
90 days
+48.67%
- Min
- $0.2600
- Avg
- $0.3137
- Max
- $0.4200
90-day median price
Specifications
The RTX 4090, Nvidia's latest high-end GPU, released in September 2022, offers significant performance and power efficiency improvements with its Ada Lovelace architecture.
- GPU Name
- AD102
- Architecture
- Ada Lovelace
- CUDA Cores / GPU
- 16384
- CUDA Cores / SM
- 128
- Die Size
- 608.5 mm2
- Frame Buffer Memory Size and Type
- 24 GB GDDR6X
- GPCs
- 11
- GPU Boost Clock (MHz)
- 2520
- GPU Codename
- AD102
- L1 Data Cache/Shared Memory
- 16384 KB
- L2 Cache Size
- 73728 KB
- Manufacturing Process
- TSMC 4N NVIDIA Custom Process
- Memory Bandwidth
- 1008 GB/sec
- Memory Clock (Data Rate)
- 21 Gbps
- Memory Interface
- 384-bit
- OFA TOPS
- 305
- PCI Express Interface
- Gen 4
- Peak BF16 TFLOPS (non-Tensor)
- 82.6
- Peak BF16 Tensor TFLOPS with FP32 Accumulate
- 165.2/330.42
- Peak FP16 TFLOPS (non-Tensor)
- 82.6
- Peak FP16 Tensor TFLOPS with FP16 Accumulate
- 330.3/660.62
- Peak FP16 Tensor TFLOPS with FP32 Accumulate
- 165.2/330.42
- Peak FP32 TFLOPS (non-Tensor)
- 82.6
- Peak FP8 Tensor TFLOPS with FP16 Accumulate
- 660.6/1321.22
- Peak FP8 Tensor TFLOPS with FP32 Accumulate
- 660.6/1321.22
- Peak INT32 TOPS (non-Tensor)
- 41.3
- Peak INT4 Tensor TOPS
- 1321.2/2642.42
- Peak INT8 Tensor TOPS
- 660.6/1321.22
- Peak TF32 Tensor TFLOPS
- 82.6/165.22
- Pixel Fill-rate (Gigapixels/sec)
- 443.5
- ROPs
- 176
- RT Cores
- 128 (3rd Gen)
- RT TFLOPS
- 191
- Register File Size
- 32768 KB
- SMs
- 128
- TGP (Total Graphics Power)
- 450 W
- TPCs
- 64
- Tensor Cores / GPU
- 512 (4th Gen)
- Tensor Cores / SM
- 4 (4th Gen)
- Texel Fill-rate (Gigatexels/sec)
- 1290.2
- Texture Units
- 512
- Transistor Count
- 76.3 Billion
- Video Engines
- 2 x NVENC (8th Gen), 1 x NVDEC (5th Gen)