Inferbase
Beta
NVIDIA A10
Datacenter
The A10 is a versatile Ampere-based GPU with 24GB GDDR6 for mainstream AI inference and graphics. Balances performance and cost for enterprise AI and virtual workstation deployments.
VRAM
24 GB
Memory
GDDR6
Bandwidth
600 GB/s
TDP
150W
Suitable Workloads
Smaller Language Models
Inference for 7B-13B parameter models
Enterprise Deployment
Designed for 24/7 datacenter operations
Key Highlights
- Built on Ampere architecture
- 288 AI accelerator cores for tensor operations
- 9,216 compute cores for parallel processing
- PCIe form factor
Memory
VRAM24 GB
Memory TypeGDDR6
Memory Bandwidth600 GB/s
Compute Performance
FP32 (Single Precision)62.5 TFLOPS
FP16 (Half Precision)125 TFLOPS
BF16125 TFLOPS
INT8250 TOPS
Architecture
ArchitectureAmpere
Compute Cores9,216
AI Accelerators288
Release Year2021
Power & Physical
TDP150 W
Max Power150 W
Form FactorPCIe
PCIe GenerationGen4 x16
Pricing
MSRP$3,500
Notes
Versatile inference GPU
Models That May Fit on NVIDIA A10
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.