NVIDIA Quadro P4000 8GB vs NVIDIA P100 PCIe 16GB

The P100 delivers 75% more FP32 throughput than the Quadro P4000 (9.3 vs 5.3 TFLOPS dense). The P100 also carries 8 GB more memory (16 GB vs 8 GB).

Quadro P4000: Pascal, 2017 P100: Pascal, 2016
FP32 dense lead
+75%
P100: 9.3 vs 5.3 TFLOPS
Memory
8 vs 16 GB
8 GB more for the P100
Bandwidth
243 GB/s vs 732 GB/s
P100 moves data faster
TDP
105 W vs 250 W
Quadro P4000 draws 145 W less

Specifications

Quadro P4000
P100
Architecture
PascalPascal
Launch Year
20172016
Form Factor
PCIePCIe
Memory
8 GB16 GB
Memory Bandwidth
243 GB/s732 GB/s
TDP
105 W250 W
Process Node
16nm

Performance (TFLOPS)

Quadro P4000
P100
FP64
No verified data available4.7 TFlops
FP32
5.3 TFlops 9.3 TFlops
TF32
No verified data available No verified data available
BF16
No verified data available No verified data available
FP16
No verified data available18.7 TFlops
FP8
No verified data available No verified data available
FP6
No verified data available No verified data available
FP4
No verified data available No verified data available
INT8
No verified data available No verified data available

FLOPS by Precision

What actually differs

The Quadro P4000 is the newer part: Pascal, launched in 2017, against the P100's Pascal from 2016. Newer architectures typically add lower-precision formats and better throughput per watt, so check the precision rows your workload actually uses.

Dense throughput for the Quadro P4000 against the P100: FP32 5.3 vs 9.3 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 8 GB against 16 GB, fed at 243 GB/s versus 732 GB/s. More memory per GPU means larger models fit before you have to shard across cards, which often matters more than raw TFLOPS for inference.

Power budgets are 105 W for the Quadro P4000 and 250 W for the P100. At FP32 that works out to 0.05 against 0.04 TFLOPS per watt.

For LLM training and inference, weight the FP8 and FP16 rows and memory capacity most heavily. For scientific and HPC workloads, the FP64 row is the one to read.

Frequently asked questions

Is the Quadro P4000 faster than the P100?

At FP32 precision the P100 reaches 9.3 TFLOPS dense against 5.3 TFLOPS for the Quadro P4000. The performance table on this page lists every published precision for both GPUs.

Which has more memory, the Quadro P4000 or the P100?

The P100 carries 16 GB of memory versus 8 GB for the Quadro P4000. Memory bandwidth is 243 GB/s for the Quadro P4000 and 732 GB/s for the P100.

How much power do the Quadro P4000 and the P100 draw?

The Quadro P4000 is rated at 105 W TDP and the P100 at 250 W. On FP32 throughput per watt, the Quadro P4000 is the more efficient part.

Can I rent the Quadro P4000 or the P100 in the cloud?

Yes. Live cloud listings tracked by Flopper start at $0.09 per GPU hour. The rental pricing section on this page lists current providers and rates for both GPUs.

Where to Rent

NVIDIA P100 PCIe 16GB

ProviderConfigurationPrice/GPU-hrChecked
Vast.ai
4× P100 PCIe Community $0.09 5h ago View →
All P100 listings and price history →

Get Comparison Updates

New GPUs added weekly. Be the first to see how they compare.