We accelerated certain AI company's new GPT open weight models -- gpt-oss-20b and gpt-oss-120b -- for leading inference performance on NVIDIA Blackwell architecture, delivering up to 1.5 million tokens per second on an NVIDIA GB200 NVL72 system.



Technical deep dive on how we
GPT0.38%
DEEP-2.41%
post-image
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • 10
  • Repost
  • Share
Comment
0/400
AirdropF5Brovip
· 08-08 18:58
The performance is off the charts!
View OriginalReply0
MetaverseMigrantvip
· 08-08 07:16
What is there to brag about with this performance?
View OriginalReply0
just_here_for_vibesvip
· 08-06 20:12
The performance improvement is so impressive.
View OriginalReply0
HAWASXvip
· 08-05 22:26
The performance improvement is so strong, the scamcoin issuance speed is about to da moon.
Reply0
MetadataExplorervip
· 08-05 22:24
NVIDIA is amazing!
View OriginalReply0
BearMarketNoodlervip
· 08-05 22:07
The performance improvement is so strong, the scamcoin issuance speed is about to da moon.
View OriginalReply0
View More
GateUser-43d6d1b5vip
· 08-05 22:06
and why doesn't the gpt increase the value?
View OriginalReply0
MEVHuntervip
· 08-05 21:59
yet another overclocked gpu eating profits... smh better optimize ur tx timing bruh
Reply0
SerumSurfervip
· 08-05 21:54
The Computing Power showdown has begun.
View OriginalReply0
View More
Trade Crypto Anywhere Anytime
qrCode
Scan to download Gate app
Community
English
  • 简体中文
  • English
  • Tiếng Việt
  • 繁體中文
  • Español
  • Русский
  • Français (Afrique)
  • Português (Portugal)
  • Bahasa Indonesia
  • 日本語
  • بالعربية
  • Українська
  • Português (Brasil)