A great American open weights MoE model that can run efficiently on your laptop or local hardware like the DGX Spark! You can use the larger Nemotron Ultra on Perplexity!

\n
@NVIDIAAI:\nIntroducing NVIDIA Nemotron 3.5 Lightning⚡
\n
An open 30B MoE model with 3B active parameters, built for always-on agents to complete high-volume, specialized tasks faster.
\n
It delivers up to 4x the output speed of similar-sized models. https://t.co/ENWrZe76pU
\n

$\"\"$

","freeContentEncoded","textContent","published","strategy","unknown","locked","deletedAt","isSyncedFromOwnBlog","original","duplicatesCount","duplicateVersion","originalArticleId","errorsCount","appearance","twitter:repost","color","backgroundColor","shortSummary","summary","bullets","persons","places","words","sentiment","subject","meta","eventId","80ac8cf25127cec36cf4508a1175da8594dd0d7a337ae92002871bd48df64eac","nostrRootEvent","event",{"_21":221,"_225":226,"_227":116,"_70":228,"_238":28,"_191":239,"_240":241},"sig","b5af04b119e5a88102c6d878d6d06ec6e76059923e0398346db98eae132bdd5a5836ed0fd81da84a03a8e2ed2cba96562050343a021cf4b5901a70f28e4d8020","kind",[229,230,234,236],[218,179],[231,232,42,233],"a","30023:983dd797fe72ad62fc8e3ece7eed1f126550972885c8f835d9af5528dad655a3:019ff4da-1374-05a2-7356-1e11e8a12fc0","mention",[235,42],"client",[237,28,42],"p","pubkey","A great American open weights MoE model that can run efficiently on your laptop or local hardware like the DGX Spark! You can use the larger Nemotron Ultra on Perplexity!\n\n\n\nnostr:naddr1qvzqqqr4gupzpxpa67tluu4dvt7gu0kw0mk37yn92ztj3pwglq6ant649rddv4drqqjrqvfevenrgerp95cnxde595cr2cfj95mnxdfk95ck2vf3v5uxzvfjve3nqj5wy49 ","created_at",1786474843,"authorNPub","markdown","A great American open weights MoE model that can run efficiently on your laptop or local hardware like the DGX Spark! You can use the larger Nemotron Ultra on Perplexity!\n\n> @NVIDIAAI:\n> Introducing NVIDIA Nemotron 3.5 Lightning⚡\n> \n> An open 30B MoE model with 3B active parameters, built for always-on agents to complete high-volume, specialized tasks faster.\n> \n> It delivers up to 4x the output speed of similar-sized models. https://t.co/ENWrZe76pU\n\n\n \n![](https://pbs.twimg.com/media/HPcUautWQAA6vS0.jpg)","freeMarkdown","kind1EventId","kind30023EventId","cdb90e10d24df0df83342b67b2c30dea32dfd352f42115120c88d0d831ecd349","kind30023",{"_21":248,"_225":251,"_227":252,"_70":253,"_238":28,"_191":265,"_240":241},"c4fff16a042dc8ffd52180a3889d44c4f6cf3cd6cebd13ff28598904cc8ffed6b3e5401a4e143ae811c091ca4ef3654dc3628582f351cb03f147659d90f737d2",30023,[254,257,258,261,263,264],[255,256],"r","https://bender.layer3.press/articles/019ff4da-1374-05a2-7356-1e11e8a12fc0",[33,179],[259,260],"published_at","1786474843",[262,174],"d",[235,42],[237,28,42],"A great American open weights MoE model that can run efficiently on your laptop or local hardware like the DGX Spark! You can use the larger Nemotron Ultra on Perplexity!\n\n> @NVIDIAAI:\n> Introducing NVIDIA Nemotron 3.5 Lightning⚡\n> \n> An open 30B MoE model with 3B active parameters, built for always-on agents to complete high-volume, specialized tasks faster.\n> \n> It delivers up to 4x the output speed of similar-sized models. https://t.co/ENWrZe76pU\n\n\n \n![](https://pbs.twimg.com/media/HPcUautWQAA6vS0.jpg)\nhttps://bender.layer3.press/articles/019ff4da-1374-05a2-7356-1e11e8a12fc0","publishedToNostr","htmlContentKind30023","pubDate","2026-08-11T19:00:43.000Z","modifiedAt","2026-08-12T07:22:40.937Z","storyId","likes","zaps","reposts","comments","tweets",[],"relatedArticles",[],"actionData","errors"]