L4

Graphics processor

Video memory: 24 GB^*
Memory bus: 192 bit^*
Rated memory bandwidth: 300.10 GB/s^*

No matched benchmark results are available for this entry yet.

Processing & platform

Architecture: Ada Lovelace^*
Process node: 5 nm^*
GPU base clock: 795 MHz^*
GPU boost clock: 2,040 MHz^*

Memory & storage

Video memory: 24 GB^*
Video memory type: Reported value 1: HBM2^*
Reported value 2: GDDR6^*
Memory bus: 192 bit^*
Rated memory bandwidth: 300.10 GB/s^*

System configuration

Detailed specifications

Specifications 1 · 18 fields

GPU architecture — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: Ada Lovelace^*
Theoretical peak: BF16 tensor dense per GPU — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: 121^* TFLOPS
Theoretical peak: FP16 tensor dense per GPU — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: 121^* TFLOPS
Theoretical peak: FP32 non-tensor dense per GPU — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: 30.3^* TFLOPS
Native hardware support: FP4 tensor — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: No^*
Native hardware support: FP6 tensor — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: No^*
Theoretical peak: FP8 tensor dense per GPU — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: 242^* TFLOPS
Theoretical peak: INT8 tensor dense per GPU — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: 242^* TOPS
Theoretical peak: TF32 tensor dense per GPU — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: 60^* TFLOPS
Physical configuration — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: PCIe, low-profile, single-slot^*
Native hardware support for GPU interconnect — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: No^*
PCIe interface declaration — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: PCIe Gen4 x16^*
Maximum board power — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: 72^* W
Peak memory bandwidth per GPU — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: 300^* GB/s
Memory interface width per GPU — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: 192^* bit
Memory capacity per GPU — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: 24^* GB
Memory technology — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: GDDR6^*
GPU model — NVIDIA L4 Tensor Core GPU — 24 GB GDDR6: NVIDIA L4 Tensor Core GPU^*

Specifications 2 · 8 fields

Product Name: L4^*
GPU Chip: AD104^*
Released: Mar 21st, 2023^*
Bus: PCIe 4.0 x16^*
Memory: 24 GB, GDDR6, 192 bit^*
GPU clock: 795^* MHz
Memory clock: 1,563^* MHz
Shaders / TMUs / ROPs: 7680 / 240 / 80^*

Specifications 3 · 58 fields

Brand: NVIDIA^*
Top BUS WIDTH: 192^* bit
Graphics Processor GPU Name: AD104^*
Graphics Processor Architecture: Ada Lovelace^*
Render Config SM Count: 60^*
Graphics Features CUDA: 8.9^*
Render Config Tensor Cores: 240^*
Graphics Processor Process Size: 5^* nm
Graphics Processor Transistors: 35,800 million^*
Graphics Processor Die Size: 294^* mm²
Graphics Card Release Date: Mar 21st, 2023^*
Graphics Card Generation: Server Ada (Lxx)^*
Graphics Card Successor: Server Hopper^*
Graphics Card Production: Active^*
Name: L4^*
Graphics Card Bus Interface: PCIe 4.0 x16^*
Clock Speeds Memory Clock: 1563 MHz 12.5 Gbps effective^*
Memory Memory Size: 24^* GB
Memory Memory Type: GDDR6^*
Memory Memory Bus: 192^* bit
Memory Bandwidth: 300.1^* GB/s
Render Config TMUs: 240^*
Top GRAPHICS PROCESSOR: AD104^*
Render Config ROPs: 80^*
Theoretical Performance Pixel Rate: 163.2 GPixel/s
Theoretical Performance Texture Rate: 489.6 GTexel/s
Board Design Slot Width: Single-slot^*
Board Design TDP: 72^* W
Board Design Suggested PSU: 250^* W
Board Design Outputs: No outputs^*
Board Design Board Number: PG193 SKU 200^*
Graphics Features DirectX: 12 Ultimate (12_2)^*
Graphics Features OpenGL: 4.6^*
Graphics Features OpenCL: 3^*
Graphics Features Vulkan: 1.3^*
source file: nvidia2023.json^*
Graphics Processor Foundry: TSMC^*
Graphics Card Predecessor: Server Ampere^*
Board Design Length: 169 mm 6.7 inches^*
Graphics Processor GPU Variant: AD104-???-A1^*
Graphics Processor Density: 121.8M / mm²^*
Board Design Height: 56 mm 2.2 inches^*
Graphics Card Reviews: 45 in our database^*
Top TMUS: 240^*
Top ROPS: 80^*
Clock Speeds Base Clock: 795^* MHz
Clock Speeds Boost Clock: 2,040^* MHz
Graphics Features Shader Model: 6.8^*
Top CORES: 7,424^*
Render Config Shading Units: 7,424^*
Render Config L2 Cache: 48^* MB
Top MEMORY SIZE: 24^* GB
Theoretical Performance FP32 (float): 30.29^* TFLOPS
Theoretical Performance FP64 (double): 473.3 GFLOPS (1:64)
Render Config L1 Cache: 128 KB (per SM)^*
Theoretical Performance FP16 (half): 30.29 TFLOPS (1:1)
Top MEMORY TYPE: GDDR6^*
Render Config RT Cores: 60^*

Specifications 4 · 14 fields

manufacturer: NVIDIA^*
rop: 128^*
igp: No^*
bus: PCIe 3.0 x16^*
memType: HBM2^*
gpuChip: GV100^*
productName: L4^*
releaseYear: 2,023^*
memSize: 24^*
memBusWidth: 192^*
gpuClock: 1,230^*
memClock: 876^*
unifiedShader: 5,120^*
tmu: 320^*

* Provisional data