Model index
OpenAI·GPT

GPT-2

Demonstrated broad task transfer from unsupervised language modeling and introduced a staged-weight release.

LegacyLanguageOpen weights
MODEL NUMBER001
First public date
2019-02-14
Organization
OpenAI
Family
GPT
01

Specifications

Only published specifications are shown.

Total parameters
1.5B
Architecture
Dense
Context
1,024 tokens
Input modalities
Text
Open weights
Yes
License
Modified MIT
Status
Legacy
02

Overview

Release dates

— OpenAI announced GPT-2 and began a staged release. ↗ — The full 1.5B GPT-2 weights were released. ↗

Why it mattered

GPT-2 brought the scale-driven emergence of general language behavior into public view—and made release governance part of the model story.

03

Specifications alongside predecessor

Compared with GPT-1 ↗

PreviousGPT-1— · 512 ctx
CurrentGPT-21.5B · 1K ctx
04

Architecture & capabilities

Model capabilities

Text generationFew-shot learning
05

Release history

Announcement

OpenAI announced GPT-2 and began a staged release.

View source
Open weights

The full 1.5B GPT-2 weights were released.

View source
06

Related models

Select a card to view the related model.

07

Sources

Suggest a correction

Copy this template, add the field, proposed value, and primary source, and share it with the archive maintainer.