OpenAI·GPT

GPT-4o

An “omni” model natively spanning text, vision, and audio with substantially lower interaction latency.

LegacyLanguageMultimodalWeights not released

MODEL NUMBER004

Specifications

Only published specifications are shown.

Architecture

Undisclosed

Context

128K

Input modalities

Text · Image · Audio

Open weights

Status

Legacy

An “omni” model natively spanning text, vision, and audio with substantially lower interaction latency.

GPT-4o moved native multimodality from a research capability into a real-time mass-market experience.

Compared with GPT-4 ↗

PreviousGPT-4— · 8K ctx

CurrentGPT-4o— · 128K ctx

New capabilities Native multimodal Real-time audio

Native multimodalVisionReal-time audio

2024-05-13General release

GPT-4o launched with native text, vision, and audio.

Select a card to view the related model.