Inferbase
Beta
NVIDIA L40
Datacenter
The L40 delivers professional visualization and AI capabilities with 48GB GDDR6 on Ada Lovelace architecture. Designed for rendering, simulation, and AI inference workloads.
VRAM
48 GB
Memory
GDDR6
Bandwidth
864 GB/s
TDP
300W
Suitable Workloads
Medium Language Models
Inference for models up to 70B parameters
Enterprise Deployment
Designed for 24/7 datacenter operations
Key Highlights
- Built on Ada Lovelace architecture
- 568 AI accelerator cores for tensor operations
- 18,176 compute cores for parallel processing
- PCIe form factor
Memory
VRAM48 GB
Memory TypeGDDR6
Memory Bandwidth864 GB/s
Compute Performance
FP32 (Single Precision)90 TFLOPS
FP16 (Half Precision)181 TFLOPS
BF16181 TFLOPS
INT8362 TOPS
Architecture
ArchitectureAda Lovelace
Compute Cores18,176
AI Accelerators568
Release Year2022
Power & Physical
TDP300 W
Max Power300 W
Form FactorPCIe
PCIe GenerationGen4 x16
Pricing
MSRP$7,000
Notes
Professional visualization
Models That May Fit on NVIDIA L40
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.