---
schema: 1
type: term
title: _GPT-4_
description: >-
  **_GPT-4_** is a multimodal AI model used in the captured material as the
  **image-to-text** step of an application: because it handles images and
  performs strong **OCR**, it transcribes a photographed book page so other
  models can interpret the text[^9Zmhe6_T-xU@211].
canonical: 'https://liquidgold.cc/channel/everyinc/explore/term/gpt-4-214b740ac94f'
markdown: 'https://liquidgold.cc/channel/everyinc/explore/term/gpt-4-214b740ac94f.md'
llms: null
parent: 'https://liquidgold.cc/channel/everyinc/llms.txt'
channel: everyinc
updated: '2026-10-23T22:12:35.346Z'
language: null
children: []
---


# \_GPT-4\_

Canonical: https://liquidgold.cc/channel/everyinc/explore/term/gpt-4-214b740ac94f

Parent: [Channel index](https://liquidgold.cc/channel/everyinc/llms.txt)

## Definition

**_GPT-4_** is a multimodal AI model used in the captured material as the **image-to-text** step of an application: because it handles images and performs strong **OCR**, it transcribes a photographed book page so other models can interpret the text[^9Zmhe6_T-xU@211].

In the captured material, **_GPT-4_** is used for one concrete job: converting an image into text. In the Hermes build — an SMS reading companion — a photo of a book page is texted in, and _GPT-4_ handles the image-to-text step because it is **multimodal** and strong at **OCR**[^9Zmhe6_T-xU@211]._GPT-4_ is deliberately paired with other models, each chosen for its strength: - **GPT-4** → image-to-text (OCR)[^9Zmhe6_T-xU@211] - **o1 preview** → interpreting the transcription, since its chain-of-thought reflection makes it better at reading text[^9Zmhe6_T-xU@234] - **o1 mini** → writing the code: smaller and faster, better at programming, though it knows less about the world[^9Zmhe6_T-xU@268] The takeaway is architectural: pick the right model for each job rather than forcing one model to do everything.- _GPT-4_ is used as the image-to-text model in the Hermes app because it is multimodal and strong at OCR.- It sits alongside o1 preview (interpretation) and o1 mini (coding), illustrating a multi-model architecture where each model handles its specialty.