Inferbase
Beta
NVIDIA T4
Datacenter
The T4 is NVIDIA's most widely deployed inference GPU with 16GB GDDR6 and 70W TDP. A cost-effective choice for inference at scale, available across all major cloud providers.
VRAM
16 GB
Memory
GDDR6
Bandwidth
320 GB/s
TDP
70W
Suitable Workloads
Smaller Language Models
Inference for 7B-13B parameter models
Enterprise Deployment
Designed for 24/7 datacenter operations
Key Highlights
- Built on Turing architecture
- 320 AI accelerator cores for tensor operations
- 2,560 compute cores for parallel processing
- PCIe Low-profile form factor
Memory
VRAM16 GB
Memory TypeGDDR6
Memory Bandwidth320 GB/s
Compute Performance
FP32 (Single Precision)8.1 TFLOPS
FP16 (Half Precision)65 TFLOPS
INT8130 TOPS
Architecture
ArchitectureTuring
Compute Cores2,560
AI Accelerators320
Release Year2018
Power & Physical
TDP70 W
Max Power70 W
Form FactorPCIe Low-profile
PCIe GenerationGen3 x16
Pricing
MSRP$2,000
Notes
Widely deployed for inference
Models That May Fit on NVIDIA T4
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.