NVIDIA RTX 6000 Ada
Workstation
VRAM
48 GB
Memory
GDDR6 ECC
Bandwidth
960 GB/s
TDP
300W
Suitable Workloads
Medium Language Models
Inference for models up to 70B parameters
Key Highlights
- Built on Ada Lovelace architecture
- 18,176 compute cores for parallel processing
- PCIe form factor
Memory
VRAM48 GB
Memory TypeGDDR6 ECC
Memory Bandwidth960 GB/s
Compute Performance
FP32 (Single Precision)91.1 TFLOPS
FP16 (Half Precision)364 TFLOPS
BF16364 TFLOPS
Architecture
ArchitectureAda Lovelace
Compute Cores18,176
Release Year2022
Power & Physical
TDP300 W
Form FactorPCIe
PCIe GenerationPCIe 4.0
Notes
FP16 dense tensor per NVIDIA datasheet. Previous-gen 48GB workstation card, still common in on-prem inference builds.
Models That May Fit on NVIDIA RTX 6000 Ada
Estimates based on INT8 quantization. Actual fit depends on framework and batch size.
Explore More
Added Jul 30, 2026
Last updated: Jul 30, 2026
Your AI stack shouldn't stand still.
Every month new models become cheaper, faster, and more capable. Inferbase ensures your application automatically benefits without changing a single API call.