NVIDIA A40
E791718
The NVIDIA A40 is a high-performance data center GPU designed for professional visualization, AI inference, and compute workloads, built on NVIDIA’s Ampere generation.
Statements (48)
| Predicate | Object |
|---|---|
| instanceOf |
NVIDIA Ampere architecture GPU
ⓘ
data center GPU ⓘ graphics processing unit ⓘ |
| architecture | Ampere NERFINISHED ⓘ |
| compatibleWith |
NVIDIA AI Enterprise software
NERFINISHED
ⓘ
NVIDIA CUDA Toolkit NERFINISHED ⓘ major deep learning frameworks ⓘ |
| cooling | passive cooling ⓘ |
| designedFor | server chassis with active airflow ⓘ |
| formFactor | PCIe add-in card ⓘ |
| hasFeature |
RT Cores
ⓘ
Tensor Cores NERFINISHED ⓘ hardware video decoding ⓘ hardware video encoding ⓘ |
| manufacturer | NVIDIA NERFINISHED ⓘ |
| memoryType | GDDR6 ⓘ |
| releaseContext | NVIDIA Ampere data center GPU family NERFINISHED ⓘ |
| supports |
AI acceleration
ⓘ
AI inference in production ⓘ ECC memory ⓘ FP16 compute ⓘ FP32 compute ⓘ INT4 compute ⓘ INT8 compute ⓘ NVIDIA CUDA NERFINISHED ⓘ NVIDIA GPU Direct NERFINISHED ⓘ NVIDIA NVDEC NERFINISHED ⓘ NVIDIA NVENC NERFINISHED ⓘ NVIDIA RTX Virtual Workstation NERFINISHED ⓘ NVIDIA RTX technology NERFINISHED ⓘ NVIDIA TensorRT NERFINISHED ⓘ NVIDIA Virtual Compute Server NERFINISHED ⓘ NVIDIA vGPU software NERFINISHED ⓘ PCI Express 4.0 NERFINISHED ⓘ containerized workloads ⓘ digital twin applications ⓘ graphics workloads ⓘ mixed-precision compute ⓘ multi-GPU configurations ⓘ ray tracing ⓘ remote visualization ⓘ rendering workloads ⓘ simulation workloads ⓘ virtualization ⓘ |
| targetMarket |
AI inference workloads
ⓘ
compute workloads ⓘ data centers ⓘ professional visualization ⓘ |
Referenced by (1)
Full triples — surface form annotated when it differs from this entity's canonical label.