Inferbase
Beta
NVIDIA L4
Datacenter
The L4 is a low-profile, energy-efficient inference accelerator with 24GB GDDR6 drawing just 72W. Ideal for high-density inference deployments in space and power-constrained environments.
VRAM
24 GB
Memory
GDDR6
Bandwidth
300 GB/s
TDP
72W
Suitable Workloads
Smaller Language Models
Inference for 7B-13B parameter models
Enterprise Deployment
Designed for 24/7 datacenter operations
Key Highlights
- Built on Ada Lovelace architecture
- 232 AI accelerator cores for tensor operations
- 7,424 compute cores for parallel processing
- PCIe Low-profile form factor
Memory
VRAM24 GB
Memory TypeGDDR6
Memory Bandwidth300 GB/s
Compute Performance
FP32 (Single Precision)30 TFLOPS
FP16 (Half Precision)121 TFLOPS
BF16121 TFLOPS
INT8242 TOPS
Architecture
ArchitectureAda Lovelace
Compute Cores7,424
AI Accelerators232
Release Year2023
Power & Physical
TDP72 W
Max Power72 W
Form FactorPCIe Low-profile
PCIe GenerationGen4 x8
Pricing
MSRP$2,500
Notes
Low-power inference
Models That May Fit on NVIDIA L4
Estimates based on INT8 quantization. Actual fit depends on framework and batch size.
Related Tools
Added Jan 25, 2026
Last updated: Jan 25, 2026
Build Your AI Stack with Confidence
Explore models, compare pricing and benchmarks, and right-size your infrastructure — all in one place.