·HSK 4

OpenAI (AI company)
OpenAI
OpenAI
to release / to launch
fābù
发布
GPT-5.5 (AI language model)
GPT-5.5
GPT-5.5
，
describing it as / calling it
chēng qí
称其
as / to be
wéi
为
up to now / to date
qìjīn
迄今
the most capable / strongest
zuì qiáng
最强
modifier particle
de
的
model
móxíng
模型
。

OpenAI (AI company)

OpenAI

on / at (time preposition)

yú

于

April 23

4 yuè 23 rì

4月23日

to launch / to release

tuīchū

推出

GPT-5.5 (AI language model)

GPT-5.5

，

this

zhè

这

shì

是

the said / this

gāi

该

series / product line

xìliè

系列

since

zì

自

GPT-4.5 (previous AI model)

GPT-4.5

since then / ever since

yǐlái

以来

the first

shǒu gè

首个

completely / fully

wánquán

完全

again / anew

chóngxīn

重新

to train

xùnliàn

训练

modifier particle

的

foundational / base

jīchǔ

基础

model

móxíng

模型

。

April 26, 2026 at 10:00 AM

OpenAI (AI company)

OpenAI

on / at (time preposition)

yú

于

this Thursday

běn zhōu sì

本周四

officially / formally

zhèngshì

正式

to launch / to release

tuīchū

推出

completion marker

了

the latest / newest

zuìxīn

最新

artificial intelligence

réngōng zhìnéng

人工智能

model

móxíng

模型

GPT-5.5 (AI language model)

GPT-5.5

。

this

zhè

这

shì

是

the said / this

gāi

该

series

xìliè

系列

among / within

zhōng

中

since

zì

自

GPT-4.5 (previous AI model)

GPT-4.5

since then / ever since

yǐlái

以来

the first

shǒu gè

首个

completely / fully

wánquán

完全

again / anew

chóngxīn

重新

to train

xùnliàn

训练

modifier particle

的

foundational / base

jīchǔ

基础

model

móxíng

模型

，

architecture / framework

jiàgòu

架构

and

hé

和

training

xùnliàn

训练

data

shùjù

数据

all / both

jūn

均

through / after going through

jīngguò

经过

comprehensive / thorough

quánmiàn

全面

update / renewal

gēngxīn

更新

。

GPT-5.5 (AI model)

GPT-5.5

in / regarding

zài

在

code / programming code

dàimǎ

代码

to write / to compose

biānxiě

编写

、

online

zàixiàn

在线

research

yánjiū

研究

and

hé

和

data

shùjù

数据

analysis

fēnxī

分析

in terms of / aspect

fāngmiàn

方面

all

jūn

均

to perform / performance

biǎoxiàn

表现

outstanding / excellent

chūsè

出色

。

new

xīn

新

model

móxíng

模型

to support

zhīchí

支持

one million

yī bǎi wàn

一百万

token (unit of text for AI)

cí yuán

词元

modifier particle

的

extra-long / extended

chāo cháng

超长

context

shàngxiàwén

上下文

window / context window

chuāngkǒu

窗口

，

to process / to handle

chǔlǐ

处理

complex / complicated

fùzá

复杂

document

wéndàng

文档

modifier particle

的

capability / ability

nénglì

能力

significantly / by a large margin

dà fú

大幅

to improve / enhance

tíshēng

提升

。

paid / fee-paying

fùfèi

付费

subscription

dìngyuè

user / subscriber

yònghù

用户

can / may

kě

可

through / via

tōngguò

通过

ChatGPT (AI chat platform)

ChatGPT

to use

shǐyòng

使用

the said / this

gāi

该

model

móxíng

模型

，

API (application programming interface)

API

input

shūrù

输入

price / cost

jiàgé

价格

is / amounts to

wéi

为

per million tokens

měi bǎi wàn cí yuán

每百万词元

$5 US dollars

5 měiyuán

5美元

，

output

shūchū

输出

is / amounts to

wéi

为

$30 US dollars

30 měiyuán

30美元

。

OpenAI (AI company)

OpenAI

to state / to indicate

biǎoshì

表示

，

GPT-5.5 (AI model)

GPT-5.5

to be able to / can

nénggòu

能够

even more

gèng

更

autonomously / independently

zìzhǔ de

自主地

to complete / accomplish

wánchéng

完成

from

cóng

从

start / beginning

kāishǐ

开始

to / until

dào

到

end / finish

jiéshù

结束

modifier particle

的

complex / complicated

fùzá

复杂

work / job

gōngzuò

工作

task / assignment

rènwù

任务

。

this time / this occasion

cǐ cì

此次

release / launch

fābù

发布

from / since (in time)

jù

距

GPT-5.4 (previous AI model)

GPT-5.4

only / merely

jǐn

仅

six weeks

liù zhōu

六周

，

to show / to indicate

xiǎnshì

显示

AI (artificial intelligence)

large language model / large AI model

dà móxíng

大模型

iteration / to iterate

diédài

迭代

speed / pace

sùdù

速度

continuously / to continue

chíxù

持续

to accelerate / to speed up

jiākuài

加快

。

New Words

OpenAI

OpenAI (AI company)

móxíng

模型

model

xùnliàn

训练

training

jiàgòu

架构

architecture

dàimǎ biānxiě

代码编写

code writing

yánjiū

研究

research

shùjù fēnxī

数据分析

data analysis

shàngxiàwén

上下文

context

dìngyuè

subscription

diédài sùdù

迭代速度

iteration speed

News in English

Science & Technology

OpenAI Releases GPT-5.5, Calling It the Most Capable Model Yet

OpenAI launched GPT-5.5 on April 23, the first fully retrained base model in the series since GPT-4.5, with enhanced agentic capabilities.

OpenAI officially launched its latest artificial intelligence model GPT-5.5 this Thursday.

This is the first completely retrained foundational model in the series since GPT-4.5, with fully updated architecture and training data.

GPT-5.5 demonstrates outstanding performance in code writing, online research, and data analysis.

The new model supports an ultra-long context window of one million tokens, with significantly enhanced ability to process complex documents.

Paid subscribers can access the model through ChatGPT, with API input pricing at $5 per million tokens and output at $30 per million tokens.

OpenAI stated that GPT-5.5 can more autonomously complete complex work tasks from start to finish.

This release comes just six weeks after GPT-5.4, showing that AI large model iteration speed continues to accelerate.

OpenAI (AI company)OpenAIOpenAIto release / to launchfābù发布GPT-5.5 (AI language model)GPT-5.5GPT-5.5，describing it as / calling itchēng qí称其as / to bewéi为up to now / to dateqìjīn迄今the most capable / strongestzuì qiáng最强modifier particlede的modelmóxíng模型。

New Words

News in English

OpenAI Releases GPT-5.5, Calling It the Most Capable Model Yet

More in Science & Technology

Moonshot AI (Chinese AI startup)Yuè zhī àn miàn月之暗面to release / to publishfābù发布global / worldwidequánqiú全球the largestzuìdà最大open-weight (AI model weights)kāiyuán quánzhòng开源权重model (AI)móxíng模型Kimi K3 (AI model by Moonshot AI)Kimi K3Kimi K3。