---
title: openai/clip-vit-large-patch14
description: "Disclaimer: The model card is taken and modified from the official CLIP repository, it can be found here.. CLIP, 428M parameters."
canonical_url: https://superlinked.com/models/openai-clip-vit-large-patch14
last_updated: 2026-05-25
---

# openai/clip-vit-large-patch14

Disclaimer: The model card is taken and modified from the official CLIP repository, it can be found here.

Source: [openai/clip-vit-large-patch14 on HuggingFace](https://huggingface.co/openai/clip-vit-large-patch14)

## Overview

| Field | Value |
|-------|-------|
| Architecture | CLIP |
| Parameters | 428M |
| Tasks | Encode |
| Outputs | Dense |
| Dimensions | Dense: 768 |
| Max sequence length | 77 tokens |
| Inputs | text, image |

## Benchmarks

### Flickr30kI2TRetrieval

Domain: general · Task: retrieval · Language: en

Image-to-text retrieval: retrieve captions from images

Corpus: 31,783 · Queries: 1,000

**Quality:** ndcg at 10: 0.7824 · map at 10: 0.6816 · mrr at 10: 0.9111

**Performance (L4 b1 c16):** Corpus 977 tok/s · Corpus p50 228.0ms · Query 7.6 mpix/s · Query p50 340.1ms

[Reference](https://shannon.cs.illinois.edu/DenotationGraph/)