Tesla P4 8GB PCIe

Specs, explained

Used for AI inference and video transcoding

1. 018 GB of GDDR5 memory

   VRAM

   8 GB

   Memory type

   GDDR5

   Memory bandwidth

   0.192 TB/s

   GDDR is graphics memory mounted on the board around the GPU rather than stacked on the package. It is widely used on graphics, rendering and cost-efficient inference cards.

   VRAM is the working space on the card. Model weights, activations and datasets have to fit in it, or be split across several GPUs. Rule of thumb for the weights alone at 16-bit precision (about 2 GB per billion parameters): up to roughly 4 billion parameters, before activations and cache. This is arithmetic on the recorded VRAM, not a benchmark.

2. 02Pascal architecture

   Architecture

   Pascal

   Launch year

   2016

   Lifecycle status

   Legacy

   Pascal introduced NVLink to NVIDIA's data-center GPUs, first on the Tesla P100. (General context about the architecture, not a statement about this specific card.)

   Launched in 2016. Kovara's records list it as a legacy part.

3. 03Form factor and power

   NVLink

   Not supported

   MIG

   Not supported

   Form factor

   PCIe low-profile single-slot

   TDP

   75 W

   Kovara's records show no NVLink support, so multi-GPU jobs on this part rely on PCIe or the network between servers to move data.

   PCIe cards slot into standard server and workstation expansion slots, which suits a wide range of systems.

   TDP is the rated thermal design power: the planning figure for power delivery and cooling in a server.

4. 04What it is used for

   Primary use (recorded)

   AI inference / video transcoding

   These are the uses Kovara has recorded for this part. The notes explain the workload in general terms; they are not performance claims.

   • AI inferenceRunning a trained model to answer requests. Memory capacity, latency and cost per request are the usual concerns.
   • video transcodingVideo transcoding and video analytics pipelines.

Figures are Kovara master data, last verified 03 Sept 2026. Explanations describe the technology in general; they are not benchmarks and say nothing about price or availability.