汉字News

← 汉字NewsChinese News for Learners
technology
kē jì
科技
·HSK 3

DeepSeek (Chinese AI company)
DeepSeek
DeepSeek
to release / to publish
fābù
发布
next-generation / new generation
xīn yī dài
新一代
flagship
qíjiàn
旗舰
AI / artificial intelligence
AI
AI
model (AI/ML)
móxíng
模型
V4 (version 4)
V4
V4
。

China
Zhōngguó
中国
AI / artificial intelligence
AI
AI
company
gōngsī
公司
DeepSeek (Chinese AI company)
DeepSeek
DeepSeek
on / at (formal preposition)
yú
于
April 24th
sì yuè èrshísì rì
四月二十四日
to release
fābù
发布
completion particle
le
了
V4 (version 4)
V4
V4
series / lineup
xìliè
系列
model
móxíng
模型
，
including
bāokuò
包括
V4-Pro (high-performance variant)
V4-Pro
V4-Pro
and
hé
和
V4-Flash (lightweight variant)
V4-Flash
V4-Flash
two
liǎng gè
两个
version
bǎnběn
版本
，
both / all
jūn
均
to support
zhīchí
支持
one million
yī bǎi wàn
一百万
token (AI/computing)
lìngpái
令牌
structural particle
de
的
extra-long / extremely long
chāo cháng
超长
context window (computing)
shàngxiàwén
上下文
。
new model
xīn móxíng
新模型
using / with
yǐ
以
open-source
kāiyuán
开源
method / approach
fāngshì
方式
to release
fābù
发布
，
pricing
dìngjià
定价
far lower than
yuǎn dī yú
远低于
competitors / rival companies
jìngzhēng duìshǒu
竞争对手
。

April 26, 2026 at 8:00 AM

China
Zhōngguó
中国
AI / artificial intelligence
AI
AI
company
gōngsī
公司
DeepSeek (Chinese AI company)
DeepSeek
DeepSeek
on / at (formal preposition)
yú
于
April 24th
sì yuè èrshísì rì
四月二十四日
officially / formally
zhèngshì
正式
to release
fābù
发布
completion particle
le
了
latest / most recent
zuì xīn
最新
flagship
qíjiàn
旗舰
model
móxíng
模型
V4 (version 4)
V4
V4
series / lineup
xìliè
系列
。
V4 (version 4)
V4
V4
series
xìliè
系列
to be divided into
fēn wéi
分为
V4-Pro (high-performance variant)
V4-Pro
V4-Pro
and
hé
和
V4-Flash (lightweight variant)
V4-Flash
V4-Flash
two
liǎng gè
两个
version
bǎnběn
版本
。
V4-Pro
V4-Pro
V4-Pro
to have a total of
gòng yǒu
共有
1.6 trillion
yī diǎn liù wànyì
一点六万亿
measure word
gè
个
parameters
cānshù
参数
，
can / able to
kě
可
to process / to handle
chǔlǐ
处理
one million
yī bǎi wàn
一百万
measure word
gè
个
token (AI/computing)
lìngpái
令牌
structural particle
de
的
extra-long
chāo cháng
超长
context window
shàngxiàwén
上下文
。
DeepSeek (Chinese AI company)
DeepSeek
DeepSeek
to state / to indicate
biǎoshì
表示
，
new model
xīn móxíng
新模型
in / at (preposition)
zài
在
programming / coding
biānchéng
编程
and
hé
和
logical reasoning
luójí tuīlǐ
逻辑推理
in terms of / aspect
fāngmiàn
方面
to have
yǒu
有
significant / great
hěn dà
很大
improvement / enhancement
tíshēng
提升
。
two types / both models
liǎng kuǎn
两款
model
móxíng
模型
both / all
dōu
都
using / with
yǐ
以
open-source
kāiyuán
开源
method / approach
fāngshì
方式
to release
fābù
发布
，
developer
kāifāzhě
开发者
can / may
kěyǐ
可以
free of charge
miǎnfèi
免费
to download
xiàzài
下载
to use
shǐyòng
使用
。
V4-Flash (lightweight variant)
V4-Flash
V4-Flash
structural particle
de
的
API (application programming interface)
API
API
pricing
dìngjià
定价
far lower than
yuǎn dī yú
远低于
competitors / rival companies
jìngzhēng duìshǒu
竞争对手
，
is
shì
是
currently / at present
mùqián
目前
most
zuì
最
cheap / affordable
piányí
便宜
structural particle
de
的
frontier / cutting-edge
qiányán
前沿
model
móxíng
模型
one of
zhī yī
之一
。
this
zhè
这
is
shì
是
DeepSeek
DeepSeek
DeepSeek
to launch / to introduce
tuīchū
推出
V3 (version 3)
V3
V3
after / following
zhī hòu
之后
structural particle
de
的
yet another / once again
yòu yī cì
又一次
significant / major
zhòngdà
重大
product
chǎnpǐn
产品
update / upgrade
gēngxīn
更新
。

New Words

DeepSeek
DeepSeek
DeepSeek (Chinese AI company)
qíjiàn móxíng
旗舰模型
flagship model
bǎnběn
版本
version
cānshù
参数
parameters
shàngxiàwén
上下文
context window
biānchéng
编程
programming
luójí tuīlǐ
逻辑推理
logical reasoning
kāiyuán
开源
open-source
dìngjià
定价
pricing
gēngxīn
更新
update / upgrade

News in English

Science & Technology

DeepSeek Releases Next-Generation Flagship AI Model V4

Chinese AI company DeepSeek released its V4 model series on April 24, featuring V4-Pro and V4-Flash variants that both support a one-million-token context window. The models were released as open-source, with pricing far below rival companies.

Chinese AI company DeepSeek officially released its latest flagship model V4 series on April 24.

The V4 series is divided into two versions: V4-Pro and V4-Flash.

V4-Pro has a total of 1.6 trillion parameters and can handle a context window of one million tokens for ultra-long context processing.

DeepSeek stated that the new model has significant improvements in programming and logical reasoning.

Both models are released in open-source form, allowing developers to download and use them for free.

V4-Flash's API pricing is far lower than competitors, making it one of the most affordable cutting-edge models currently available.

This represents another major product update from DeepSeek following the release of V3.

More in Science & Technology

MiniMax
MiniMax
MiniMax
releases
fābu
发布
M3
M3
M3
multimodal
duōmótài
多模态
model
móxíng
模型
。

MiniMax M3 Multimodal Model Delivers 9x Faster Performance

Nvidia
yīngwěidá
英伟达
releases
fābu
发布
Cosmos
Cosmos
Cosmos
3
3
3
open-source
kāiyuán
开源
physical
wùlǐ
物理
AI
réngōng zhìnéng
人工智能
model
móxíng
模型
。

Nvidia Releases Cosmos 3 Open-Source Physical AI Model

Apple (company)
Píngguǒ
苹果
to release / to launch
fābù
发布
brand new / completely new
quánxīn
全新
Siri (Apple voice assistant)
Siri
Siri
，
features / functions
gōngnéng
功能
greatly / significantly
dàfú
大幅
to upgrade / upgraded
shēngjí
升级
。

Apple Unveils Completely Rebuilt Siri at WWDC 2026

Germany
Déguó
德国
robot / robotics
jīqìrén
机器人
company
gōngsī
公司
Neura Robotics (German humanoid robot company)
Neura Robotics
Neura Robotics
to complete / to close
wánchéng
完成
$1.4 billion
14 yì měiyuán
14亿美元
funding round
róngzī
融资
。

German Robotics Firm Neura Robotics Closes $1.4 Billion Funding Round