{"id":38402,"date":"2025-07-26T09:20:14","date_gmt":"2025-07-26T09:20:14","guid":{"rendered":"https:\/\/www.newsbeep.com\/us\/38402\/"},"modified":"2025-07-26T09:20:14","modified_gmt":"2025-07-26T09:20:14","slug":"build-more-accurate-and-efficient-ai-agents-with-the-new-nvidia-llama-nemotron-super-v1-5","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/us\/38402\/","title":{"rendered":"Build More Accurate and Efficient AI Agents with the New NVIDIA Llama Nemotron Super v1.5"},"content":{"rendered":"<p>The <a href=\"https:\/\/www.nvidia.com\/en-us\/ai-data-science\/foundation-models\/nemotron\/\" data-wpel-link=\"internal\" target=\"_self\" rel=\"follow noopener nofollow\">NVIDIA Nemotron <\/a>family builds on the strongest open models in the ecosystem by enhancing them with greater accuracy, efficiency, and transparency using NVIDIA open synthetic datasets, advanced techniques, and tools.<\/p>\n<p>Today, we\u2019re introducing NVIDIA Llama Nemotron Super v1.5, which brings significant improvements across core reasoning and agentic tasks like math, science, coding, function calling, instruction following, and chat, while maintaining strong throughput and compute efficiency.<\/p>\n<p>Built for reasoning and agentic workloads<a href=\"#built_for_reasoning_and_agentic_workloads\" class=\"heading-anchor-link\"><\/a><\/p>\n<p>Llama Nemotron Super v1.5 builds on the same efficient reasoning foundation as Llama Nemotron Ultra. However, the model has been refined through post-training using a new dataset focused specifically on high-signal reasoning tasks.<\/p>\n<p>Across a wide range of benchmarks, Llama Nemotron Super v1.5 outperforms other open models in its weight class, particularly in tasks that require multi-step reasoning and structured tool use.<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" width=\"2466\" height=\"1120\" src=\"https:\/\/www.newsbeep.com\/us\/wp-content\/uploads\/2025\/07\/image-29-png.webp.webp\" alt=\"The bar graph shows the leading accuracy comparison of popular leading models and Llama Nemotron Super v1.5 delivers the highest accuracy for reasoning and agentic tasks.\" class=\"lazyload wp-image-103844\"  data-\/>Figure 1. Llama Nemotron Super v1.5 delivers the highest accuracy for reasoning and agentic tasks<\/p>\n<p>To boost throughput and deployment efficiency, we applied pruning techniques like neural architecture search. Higher throughput means the model can reason faster and explore more complex problem spaces within the same compute and time budget\u2014delivering stronger reasoning at lower inference costs. It also runs on a single GPU, further reducing compute overhead.<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" width=\"997\" height=\"501\" src=\"https:\/\/www.newsbeep.com\/us\/wp-content\/uploads\/2025\/07\/1753521614_308_image-7.png\" alt=\"The bar chart compares token generation for Llama Nemotron models. The Llama Nemotron Super model delivers the highest throughput.\" class=\"lazyload wp-image-103847\"  data-\/>Figure 2. Llama Nemotron Super v1.5 provides the highest accuracy and throughput for agentic tasks, lowering the cost of inference<\/p>\n<p>Try the model now<a href=\"#try_the_model_now\" class=\"heading-anchor-link\"><\/a><\/p>\n<p>Experience Llama Nemotron Super v1.5 now at <a href=\"https:\/\/build.nvidia.com\/nvidia\/llama-3_3-nemotron-super-49b-v1_5\" data-wpel-link=\"internal\" target=\"_self\" rel=\"follow noopener nofollow\">build.nvidia.com<\/a>, or download the model directly from <a href=\"https:\/\/huggingface.co\/nvidia\/Llama-3_3-Nemotron-Super-49B-v1_5\" data-wpel-link=\"external\" target=\"_blank\" rel=\"follow external noopener nofollow\">Hugging Face<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"The NVIDIA Nemotron family builds on the strongest open models in the ecosystem by enhancing them with greater&hellip;\n","protected":false},"author":2,"featured_media":38403,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[45],"tags":[182,181,507,74],"class_list":["post-38402","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-artificial-intelligence","tag-artificialintelligence","tag-technology"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/38402","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/comments?post=38402"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/38402\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media\/38403"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media?parent=38402"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/categories?post=38402"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/tags?post=38402"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}