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•5m agoBase Leader: x402 now supports batch settlement
ChainCatcher•13m agoAnthropic is in talks to raise at least $30 billion at a $900 billion valuation
Odaily•24m agoSpark launches the 3rd month SPK buyback
ChainCatcher•45m agoData: A wallet that has been dormant for a year spent 5.81 million USDC to purchase 2,570 ETH 7 hours ago
ChainCatcher•3h agoJPMorgan Chase will launch a second tokenized money market fund on Ethereum
ChainCatcher•3h agoData: Bitcoin's market share has rebounded to more than 58%, and the market may enter a consolidation stage
TechFlow•8h agoAave and Kelp have completed the first steps of the rsETH technical recovery plan and have burned the attacker's rsETH on Arbitrum
Odaily•9h agoThe U.S. Senate has received enough votes to confirm Walsh's appointment as a member of the Federal Reserve Board
TechFlow•9h agoHackers implant malicious code in the Mistral AI package
ChainCatcher•9h agoThe Ethereum Working Group has launched a clear signature open standard to solve the problem of blind signatures and improve the security of on-chain transactions