NVIDIA推出Nemotron 3 Nano Omni模型，多模态推理效率提升9倍
Odaily星球日报讯 NVIDIA 在 X 平台发文表示，今日推出开源多模态模型 Nemotron 3 Nano Omni。该模型采用 30B-A3B 混合专家架构（MoE），支持 256K 上下文，可统一处理视频、音频、图像及文本输入。与同类交互水平的开源全能模型相比，该模型吞吐量提升达 9 倍，大幅降低推理成本并提高扩展性。目前 Nemotron 3 Nano Omni 已在 Hugging Face、OpenRouter 及 NVIDIA NIM 上线，并获得 Aible、Applied Scientific Intelligence、H Company 等企业采用。
Disclaimer: OKX Orbit content is provided for informational purposes only. Learn more
Replies
Related Flash News
Odaily•19m agoIEA: Oil prices could soar further as global oil inventories fall at a record pace
ChainCatcher•46m agoCryptoQuant: Altcoins are showing signs of recovery, but it's still too early to tell if the "altcoin season" has begun
ChainCatcher•1h ago数据：KULR Technology 向 Coinbase Prime 转入 300 枚 BTC，浮亏 1825 万美元
Odaily•1h agoSoftBank: The cumulative book income on OpenAI as of the end of March was $45 billion
ChainCatcher•2h agoAnthropic intends to acquire Stainless for at least $300 million
Odaily•2h agoBIT US stocks have a 100-day AUM of more than $200 million, and you can invest in US stocks without a bank card
ChainCatcher•3h agoThe Hong Kong High Court froze the assets of Prince Group Chen Zhi and others for HK$8.938 billion
ChainCatcher•4h agoThe 21Shares Hyperliquid ETF (THYP) recorded a solid performance with a trading volume of $1.8 million on the first day, with analysts calling it solid
ChainCatcher•4h agoData: Alameda Research withdrew more than $20 million in assets from KuCoin in the past 2 hours
Odaily•4h agoDeposit 3,000 BNB to Pionex 45 minutes before the GMGN fee related address