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
ChainCatcher•13m agoLighter completes the L2 escape mechanism verification, and users can still withdraw money to the mainnet when the sequencer fails
ChainCatcher•15m agoMultiple Bankr user wallets were stolen, involving Grok's interaction with Bankrbot resulting in unauthorized transaction signatures
ChainCatcher•52m agoGrafana: The investigation found that recent security incidents have not affected customers' production systems or operations
ChainCatcher•56m agoBank of England Deputy Governor: Tokenization can reduce costs, accelerate settlements, and promote competition
ChainCatcher•1h agoData: Whales associated with a16z purchased and staked 206,325 HYPE in their new wallets
ChainCatcher•1h agoHuatai Securities: The Federal Reserve may raise interest rates twice next year
ChainCatcher•1h agoTrump signed an executive order to review restrictions on crypto companies' access to U.S. payment systems
ChainCatcher•1h agoIrish drug dealer Bitcoin address transferred 500 BTC again
ChainCatcher•2h agoPaulson: Keep interest rates steady to assess economic conditions
ChainCatcher•2h agoPaulson: The market considers a scenario where rates remain unchanged or rise is "healthy"