Data updated 2026-11-04 15:09 UTC · refreshed every 5 min
Current Price
$0.85
-1.9% (24h)
Available Now
25 GPU
Good availability
Total in Network
89 GPU
Live data
Utilization
71.9%
Stable
Market history
USD/hour (minimum · median)
Ready to rent
Infrastructure Growth
Estimated with our VRAM calculator for a single card: 8K context, 4 concurrent users, vLLM-style serving.
35B MoE · 3B active
~65 tok/s per stream
Where the memory goes
14B
~18 tok/s per stream
Where the memory goes
4B
~54 tok/s per stream
Where the memory goes
Estimates, not benchmarks: real usage depends on the runtime, context and batch settings.
Memory
48GB GDDR6
FP32 Performance
37.4 TFLOPS
Tensor Performance
150 TFLOPS
Memory Bandwidth
696 GB/s
Architecture
Ampere
TDP
300W
FP64 Performance
0.6 TFLOPS
CUDA Cores
10752
Tensor Cores
336
The NVIDIA A40 is a versatile data center GPU for AI inference, visualization, and rendering.
The NVIDIA A40 features 48GB of GDDR6 memory with 696 GB/s bandwidth, making it excellent for ai inference, professional visualization, rendering. With 150 TFLOPS tensor performance and 37.4 TFLOPS FP32 compute, this NVIDIA Ampere GPU delivers exceptional performance for demanding workloads.
Currently, 89 GPU A40 GPUs are available in the GPUniq network with 71.9% utilization. The average rental price is $0.85 per hour, with 25 GPU ready for immediate deployment.
Prices are live minimum / median USD per hour on the GPUniq marketplace, as of 2026-11-04 15:09 UTC.
The current rental price for NVIDIA A40 is $0.85 per hour on GPUniq. Prices may vary based on demand and availability.
NVIDIA A40 is ideal for AI Inference, Professional Visualization, Rendering, Data Science, Virtual Workstation. With 48GB of GDDR6 memory and 150 TFLOPS tensor performance, it excels at demanding AI and compute workloads.
Currently, there are 25 GPU available for immediate rental out of 89 GPU total in the GPUniq network. The current utilization is 71.9%.
On a single NVIDIA A40, our VRAM calculator fits Qwen3.6 35B A3B (INT4 / AWQ, 23.8 of 48 GB); Qwen3-14B (FP16, 35.6 of 48 GB); Qwen3.5-4B (FP16, 11.4 of 48 GB) — estimated at 8K context with 4 concurrent users.
NVIDIA A40 features 48GB GDDR6 memory, 37.4 TFLOPS FP32 performance, 150 TFLOPS tensor performance, 696 GB/s memory bandwidth, and 300W TDP. It uses the Ampere architecture.
NVIDIA official specificationsAPI docsAbout GPUniqContact supportPrivacy policyTerms of use
$0.85/hr
A40 · 25 GPU