LLaVA vs. GPT-4 Vision: Compared and Contrasted

Models

LLaVA-1.5

LLaVA is an open source multimodal language model that you can use for visual question answering and has limited support for object detection.

Learn more about LLaVA-1.5

GPT-4 with Vision

GPT-4 with Vision is a multimodal language model developed by OpenAI.

Learn more about GPT-4 with Vision

Model Type

Object Detection

Model Features

Item 1 Info

Item 2 Info

Architecture

Transformer

Frameworks

Annotation Format

Instance Segmentation

GitHub

View Repo

GitHub Stars

16,000

License

Apache-2.0

Paper

View Paper

Training Notebook

Train on Colab

Deploy Model

Deploy with Roboflow

Compare Alternatives

Compare LLaVA-1.5 and GPT-4 with Vision with Autodistill

We ran seven tests across five state-of-the-art Large Multimodal Models (LMMs) on November 23rd, 2023. GPT-4V passed at four of seven tests and LLaVA passed at one of seven tests.

Here are the results:

Based on our tests, GPT-4V performs better than LLaVA at multimodal tasks.