# Tesla P4

> NVIDIA · Pascal

Discontinued

Canonical: https://gpuninja.com/chips/tesla-p4
Data as of: 2026-10-11T10:45:05Z
Path: Home › GPU chips

> **Discontinued.** Discontinued on 2 Dec 2025. This page is a preserved snapshot of the last recorded specs, prices and benchmark results.

- Specs verified: between 11 Oct 2026, 08:44 UTC and 11 Oct 2026, 10:45 UTC

## Specifications

- Vendor: NVIDIA (Verified · source nvidia.com · retrieved 11 Oct 2026, 06:35 UTC · checked 11 Oct 2026, 08:44 UTC)
- Architecture: Pascal (Verified · source nvidia.com · retrieved 11 Oct 2026, 06:35 UTC · checked 11 Oct 2026, 08:44 UTC)
- Die: GP104 (Verified · source nvidia.com · retrieved 11 Oct 2026, 06:35 UTC · checked 11 Oct 2026, 08:44 UTC)
- Process: TSMC 16 nm (Verified · source en.wikipedia.org · retrieved 11 Oct 2026, 06:30 UTC · checked 11 Oct 2026, 08:44 UTC)
- Shader cores: 2,560 (Verified · source nvidia.com · retrieved 11 Oct 2026, 06:35 UTC · checked 11 Oct 2026, 08:44 UTC)
- Tensor cores: None
- RT cores: None
- Memory bus: 256-bit (Verified · source nvidia.com · retrieved 11 Oct 2026, 06:35 UTC · checked 11 Oct 2026, 08:44 UTC)
- Max VRAM: 8 GB (Verified · source nvidia.com · retrieved 11 Oct 2026, 06:35 UTC · checked 11 Oct 2026, 08:44 UTC)
- Reference boost clock: 1,531 MHz (Verified · source nvidia.com · retrieved 11 Oct 2026, 06:35 UTC · checked 11 Oct 2026, 08:44 UTC)
- Reference TDP: 75 W (Verified · source nvidia.com · retrieved 11 Oct 2026, 06:35 UTC · checked 11 Oct 2026, 08:44 UTC)
- Host interface: PCIe 3.0 x16 (Verified · source nvidia.com · retrieved 11 Oct 2026, 10:07 UTC · checked 11 Oct 2026, 10:45 UTC)
- Launch MSRP: No public MSRP
- Launch date: 12 Sep 2016 (Verified · source nvidianews.nvidia.com · retrieved 11 Oct 2026, 06:40 UTC · checked 11 Oct 2026, 08:44 UTC)
- Segment: Data center (Verified · source nvidia.com · retrieved 11 Oct 2026, 06:35 UTC · checked 11 Oct 2026, 08:44 UTC)
- PCI IDs: 10de:1bb3 (Verified · source nvidia.com · retrieved 11 Oct 2026, 06:35 UTC · checked 11 Oct 2026, 08:44 UTC)
- Status: Discontinued on 2 Dec 2025 (Verified · source download.nvidia.com · retrieved 11 Oct 2026, 06:46 UTC · checked 11 Oct 2026, 08:44 UTC)

## Performance

- Memory bandwidth: 192 GB/s (Verified · source nvidia.com · retrieved 11 Oct 2026, 10:07 UTC · checked 11 Oct 2026, 10:45 UTC)
- FP32 compute: 5.5 TFLOPS (Verified · source nvidianews.nvidia.com · retrieved 11 Oct 2026, 10:09 UTC · checked 11 Oct 2026, 10:45 UTC)
- FP16 compute: 0.09 TFLOPS (Unverified · source docs.nvidia.com · retrieved 11 Oct 2026, 10:08 UTC · checked 11 Oct 2026, 10:45 UTC)
- BF16 compute: Not supported
- FP8 compute: Not supported
- FP4 compute: Not supported
- INT8 compute: 22 TOPS (Verified · source nvidianews.nvidia.com · retrieved 11 Oct 2026, 10:09 UTC · checked 11 Oct 2026, 10:45 UTC)

## Boards built on this chip

Board, Chip, VRAM, TDP, Length, Slots, Price, Stock

No boards recorded for this chip.

## Benchmarks

No benchmark results recorded.

## Sources

- https://www.nvidia.com/content/dam/en-zz/Solutions/design-visualization/solutions/resources/documents1/Tesla-P4-Product-Brief.pdf — Vendor, Architecture, Die, Shader cores, Memory bus, Max VRAM, Reference boost clock, Reference TDP, Host interface, Segment, PCI IDs, Memory bandwidth; last verified 11 Oct 2026, 10:45 UTC
- https://en.wikipedia.org/wiki/Pascal_(microarchitecture) — Process; last verified 11 Oct 2026, 08:44 UTC
- https://nvidianews.nvidia.com/news/new-nvidia-pascal-gpus-accelerate-deep-learning-inference — Launch date, FP32 compute, INT8 compute; last verified 11 Oct 2026, 10:45 UTC
- https://download.nvidia.com/XFree86/Linux-x86_64/590.44.01/README/supportedchips.html — Status; last verified 11 Oct 2026, 08:44 UTC
- https://docs.nvidia.com/cuda/archive/12.4.0/cuda-c-programming-guide/index.html — FP16 compute; last verified 11 Oct 2026, 10:45 UTC