NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB, Advancing Agentic Retrieval NVIDIA Nemotron 3 Embed 在 RTEB 榜单中荣登榜首，推动智能体检索技术发展

Retrieval is critical in multi-step agentic workflows where poor retrieval can cause agents to fetch irrelevant context, re-query, waste token budget, and carry noise into later reasoning steps.

Today, we are releasing NVIDIA Nemotron 3 Embed, a collection of open and commercially available embedding models designed to improve retrieval quality while giving developers practical deployment options for production-scale RAG, agentic retrieval, code retrieval, and agent memory.今天，我们发布了 NVIDIA Nemotron 3 Embed，这是一系列开放且可商用的嵌入（Embedding）模型。该系列旨在提升检索质量，同时为开发者提供实用的部署方案，以应对生产规模的 RAG（检索增强生成）、智能体检索、代码检索和智能体记忆等需求。

The collection includes three open models that achieve state-of-the-art retrieval across the accuracy-efficiency curve, led by an 8B model that tops the RTEB leaderboard and efficient 1B variants built for production-scale deployment:该系列包含三款开源模型，在准确率与效率的平衡曲线上均达到了业界领先水平。其中包括在 RTEB 排行榜上位居第一的 8B 模型，以及专为生产级部署打造的高效 1B 变体：

Table 1. Nemotron 3 Embed Model Usability and Deployment Matrix.表 1. Nemotron 3 Embed 模型可用性与部署矩阵。

Figure 1. RTEB Multilingual Leaderboard screenshot (July 15, 2026) showing Nemotron-3-Embed-8B-BF16 ranked as #1.图 1. RTEB 多语言排行榜截图（2026 年 7 月 15 日），显示 Nemotron-3-Embed-8B-BF16 排名第一。

Key Features 主要特性

Beyond the RTEB result, Nemotron 3 Embed introduces a production-ready feature set for enterprise retrieval deployments:除了在 RTEB 取得优异成绩外，Nemotron 3 Embed 还为企业级检索部署引入了一套生产就绪的功能集：

• Open Weights, Datasets, and Recipes: Gives teams control to inspect, tune, fine-tune, and deploy retrieval models on their own infrastructure. 开放权重、数据集和方案：赋予团队在自有基础设施上检查、调整、微调和部署检索模型的控制权。
• 32k Context Window: Supports retrieval over long documents, large code contexts, and multi-turn agent histories while reducing truncation. 32k 上下文窗口：支持对长文档、大型代码上下文和多轮智能体历史记录进行检索，同时减少截断现象。
• Multilingual & Code Retrieval: Supports retrieval across global enterprise data, technical documentation and multi-file code repositories. 多语言与代码检索：支持跨全球企业数据、技术文档和多文件代码库的检索。
• NVIDIA NVFP4 Efficiency: Provides a Blackwell-optimized 4-bit deployment path for high-throughput retrieval with a smaller memory footprint. NVIDIA NVFP4 效率：提供针对 Blackwell 架构优化的 4-bit 部署路径，以更小的内存占用实现高吞吐量检索。
• Fine-Tuning and Distillation Recipes: NVIDIA NeMo AutoModel recipes support domain adaptation and model compression for teams adapting retrieval models to their own data. 微调与蒸馏方案：NVIDIA NeMo AutoModel 方案支持领域适配和模型压缩，方便团队将检索模型应用于自有数据。
• Day-0 Ecosystem Integration: Available immediately on Hugging Face, deployable as NVIDIA NIM microservice, supported by vLLM, and accessible through leading AI Cloud and inference partners.Day-0 生态系统集成：现已在 Hugging Face 上线，可作为 NVIDIA NIM 微服务部署，支持 vLLM，并可通过领先的 AI 云和推理合作伙伴获取。

Evaluation: Retrieval Quality, Agentic Efficiency, and Deployment Tradeoffs 评估：检索质量、智能体效率与部署权衡

We evaluate Nemotron 3 Embed across three dimensions: retrieval quality, downstream agentic efficiency, and deployment tradeoffs. The 8B model establishes the model collection’s quality ceiling, while the 1B BF16 and NVFP4 variants bring the same retrieval-focused design to lower-cost and higher-throughput deployment settings.我们从检索质量、下游智能体效率和部署权衡三个维度对 Nemotron 3 Embed 进行了评估。8B 模型确立了该系列模型的质量上限，而 1B BF16 和 NVFP4 变体则将同样的检索优化设计带入了成本更低、吞吐量更高的部署场景。

RTEB Leadership and Strong Gains Across Retrieval Benchmarks RTEB 领先地位与检索基准测试的显著提升

We first evaluated the models on RTEB, where Nemotron-3-Embed-8B-BF16 ranks #1. We also tested these models across ViDoRe V3 Text, and MMTEB Retrieval and LongEmbed using average NDCG@10.我们首先在 RTEB 上评估了这些模型，其中 Nemotron-3-Embed-8B-BF16 排名第一。我们还使用平均 NDCG@10 指标，在 ViDoRe V3 Text、MMTEB Retrieval 和 LongEmbed 上对这些模型进行了测试。

Figure 2. Retrieval accuracy using average NDCG@10 across RTEB, ViDoRe V3 Text, MMTEB Retrieval and LongEmbed, comparing the Nemotron 3 Embed models with prior-generation Nemotron baselines.图 2. 使用 RTEB、ViDoRe V3 Text、MMTEB Retrieval 和 LongEmbed 的平均 NDCG@10 指标衡量检索准确率，比较了 Nemotron 3 Embed 模型与上一代 Nemotron 基准模型。

• Nemotron-3-Embed-8B-BF16 ranks #1 on RTEB, scoring 78.5% on RTEB and 75.5% on MMTEB Retrieval. Nemotron-3-Embed-8B-BF16 在 RTEB 上排名第一，RTEB 得分为 78.5%，MMTEB Retrieval 得分为 75.5%。
• Nemotron-3-Embed-1B-BF16 brings much of the 8B model’s retrieval quality into a smaller deployment footprint. It scores 72.4% on RTEB, reducing error rate by 27% over its 1B predecessor (llama-nemotron-embed-vl-1b-v2), and scores 71.0% on MMTEB Retrieval, reducing error rate by 28%.Nemotron-3-Embed-1B-BF16 将 8B 模型的大部分检索质量带入了更小的部署规模中。其 RTEB 得分为 72.4%，较其 1B 前身（llama-nemotron-embed-vl-1b-v2）降低了 27% 的错误率；在 MMTEB Retrieval 上得分为 71.0%，错误率降低了 28%。

Why Better Retrieval Matters for Agents 为何更好的检索对智能体至关重要

To evaluate retrieval in an agentic setting, we use a search agent powered by Nemotron 3 Ultra and vary the embedding model used by the retrieval system. Better retrieval can return relevant evidence earlier, helping the agent avoid repeated searches, unnecessary reasoning turns, and extra context inspection. We compare average retrieval accuracy with estimated downstream agentic token cost per query across ViDoRe V3, BRIGHT, and BrowseComp-Plus.为了评估智能体环境下的检索效果，我们使用由 Nemotron 3 Ultra 驱动的搜索智能体，并改变检索系统所使用的嵌入模型。更好的检索能够更早地返回相关证据，从而帮助智能体避免重复搜索、不必要的推理轮次以及额外的上下文检查。我们比较了在 ViDoRe V3、BRIGHT 和 BrowseComp-Plus 上，平均检索准确率与预估的下游智能体每查询 Token 成本。

Figure 3. Average retrieval accuracy versus downstream agentic token cost per query across ViDoRe V3, BRIGHT, and BrowseComp-Plus.图 3. 在 ViDoRe V3、BRIGHT 和 BrowseComp-Plus 上，平均检索准确率与下游智能体每查询 Token 成本的对比。

Evaluation note: The search agent uses Nemotron 3 Ultra. Downstream token cost is estimated from Nemotron 3 Ultra input/output token counts using the GPT-5.5 pricing formula.评估说明：搜索智能体使用 Nemotron 3 Ultra。下游 Token 成本是根据 Nemotron 3 Ultra 的输入/输出 Token 数量，并使用 GPT-5.5 定价公式估算的。

Figure 3 shows that stronger retrieval reduces downstream agentic token cost. More accurate retrievers return relevant evidence earlier, which helps agents complete tasks with fewer repeated searches and fewer reasoning turns. In these evaluations, the Nemotron 3 Embed models improve the agentic retrieval frontier, with the 8B model delivering both the highest average retrieval accuracy and the lowest estimated downstream token cost across ViDoRe V3, BRIGHT, and BrowseComp-Plus.图 3 显示，更强的检索能力可降低下游智能体的 Token 成本。更准确的检索器能更早返回相关证据，帮助智能体以更少的重复搜索和推理轮次完成任务。在这些评估中，Nemotron 3 Embed 模型提升了智能体检索的边界，其中 8B 模型在 ViDoRe V3、BRIGHT 和 BrowseComp-Plus 上均提供了最高的平均检索准确率和最低的预估下游 Token 成本。

Scaling Retrieval with NVFP4 on Blackwell 利用 Blackwell 架构上的 NVFP4 扩展检索

For high-throughput deployments, teams often choose smaller embedding models to meet latency and cost targets. Nemotron-3-Embed-1B-NVFP4 is designed to narrow the gap between serving efficiency and retrieval quality by using native NVFP4 acceleration on NVIDIA Blackwell architectures. The model quantizes the weights and activations of linear layers to NVFP4 for efficient inference, and uses Quantization-Aware Distillation (QAD) to help recover accuracy for long input sequences. 对于高吞吐量部署，团队通常会选择更小的嵌入模型以满足延迟和成本目标。Nemotron-3-Embed-1B-NVFP4 旨在通过在 NVIDIA Blackwell 架构上使用原生 NVFP4 加速，缩小服务效率与检索质量之间的差距。该模型将线性层的权重和激活值量化为 NVFP4 以实现高效推理，并使用量化感知蒸馏（QAD）技术帮助恢复长输入序列的准确性。

Figure 4. ViDoRe V3 retrieval accuracy versus serving efficiency, comparing Nemotron-3-Embed-1B-NVFP4 with selected smaller open embedding baselines, including Qwen3-Embedding-0.6B and EmbeddingGemma-300M.图 4. ViDoRe V3 检索准确率与服务效率对比，比较了 Nemotron-3-Embed-1B-NVFP4 与选定的较小开源嵌入基准模型（包括 Qwen3-Embedding-0.6B 和 EmbeddingGemma-300M）。

• Serving Efficiency: NVFP4 on Blackwell delivers up to 2x higher throughput than BF16 for high-throughput, low-latency retrieval serving. 服务效率：对于高吞吐量、低延迟的检索服务，Blackwell 架构上的 NVFP4 提供的吞吐量比 BF16 高出 2 倍。
• Accuracy retention: The NVFP4 variant retains 99%+ of BF16 retrieval accuracy while reducing memory footprint.准确率保持：NVFP4 变体在保持 99% 以上 BF16 检索准确率的同时，减少了内存占用。

Day 0 Performant NIM Day 0 高性能 NIM

For production-scale retrieval systems, the serving stack also needs to preserve that efficiency under real request loads, across different input sequence lengths and hardware targets. To make Nemotron 3 Embed performant at enterprise scale today, we are also releasing an optimized NVIDIA NIM microservice for the 1B model. As shown in Figure 5, the Rust-based Nemotron 3 Embed NIM matches or outperforms the vLLM checkpoint on NVIDIA GB200 and RTX PRO 6000 GPUs across ISLs of 256 & 1024.对于生产级检索系统，服务栈还需要在实际请求负载下、跨不同输入序列长度和硬件目标保持这种效率。为了让 Nemotron 3 Embed 在当今的企业规模下实现高性能，我们还为 1B 模型发布了优化的 NVIDIA NIM 微服务。如图 5 所示，基于 Rust 的 Nemotron 3 Embed NIM 在 NVIDIA GB200 和 RTX PRO 6000 GPU 上，针对 256 和 1024 的输入序列长度（ISL），均达到或超过了 vLLM 检查点的性能。

Figure 5. Nemotron 3 Embed NIM serving performance compared with the vLLM checkpoint on NVIDIA GB200 and RTX PRO 6000 GPUs.图 5. Nemotron 3 Embed NIM 服务性能与 NVIDIA GB200 和 RTX PRO 6000 GPU 上 vLLM 检查点的对比。

How We Built the Nemotron 3 Embed Models 我们如何构建 Nemotron 3 Embed 模型

Nemotron-3-Embed-8B-BF16 adapts the Ministral-3-8B-Instruct-2512 backbone by converting its causal decoder into a bidirectional encoder for full-sequence retrieval. The model is trained with contrastive pre-training on a blend of web-sourced and synthetic text pairs, then fine-tuned on curated multilingual retrieval datasets across domains such as legal, finance, medical, business, and education. This 8B model serves as the flagship embedding model, while earlier 8B teacher checkpoints from the same development line were used to distill the efficient 1B variants.Nemotron-3-Embed-8B-BF16 通过将 Ministral-3-8B-Instruct-2512 主干的因果解码器转换为用于全序列检索的双向编码器来进行适配。该模型在网络来源和合成文本对的混合数据上进行了对比预训练，随后在法律、金融、医疗、商业和教育等领域的精选多语言检索数据集上进行了微调。这款 8B 模型作为旗舰嵌入模型，而同一开发路线中较早的 8B 教师检查点则被用于蒸馏出高效的 1B 变体。

Scaling Down to 1B 扩展至 1B

The 1B model is not a small retriever trained from scratch. We first applied the bidirectional adaptation recipe to the Ministral-3-3B-Instruct-2512 backbone to establish a 3B retriever base, then compressed it through two-rounds of structured pruning and distillation.1B 模型并非从头开始训练的小型检索器。我们首先将双向适配方案应用于 Ministral-3-3B-Instruct-2512 主干，建立了一个 3B 检索器基座，然后通过两轮结构化剪枝和蒸馏对其进行了压缩。

First, the 3B parent model was compressed to a 2B intermediate footprint using NVIDIA ModelOpt’s mcore_minitron Neural Architecture Search engine. The NAS pipeline searched across hidden width, FFN size, attention heads, and depth under a strict parameter budget to identify an efficient architecture for retrieval workloads.首先，利用 NVIDIA ModelOpt 的 mcore_minitron 神经架构搜索（NAS）引擎，将 3B 父模型压缩至 2B 中间规模。NAS 流水线在严格的参数预算下，对隐藏层宽度、FFN 大小、注意力头数和深度进行了搜索，以识别出适合检索工作负载的高效架构。

The resulting 2B intermediate model was then distilled from an 8B teacher checkpoint to recover ranking accuracy. We used a combined cosine distance loss and mean squared error loss on a multilingual, in-domain retrieval data blend to align the student’s embeddings with the teacher.随后，将所得的 2B 中间模型从 8B 教师检查点进行蒸馏，以恢复排序准确性。我们在多语言、领域内检索数据混合集上使用了组合余弦距离损失和均方误差损失，将学生的嵌入与教师模型对齐。

Figure 6. The pruning and distillation pipeline compresses the retriever from a 3B base to the final 1B production model.图 6. 剪枝和蒸馏流水线将检索器从 3B 基座压缩至最终的 1B 生产模型。

This same sequence, ModelOpt structured pruning followed by 8B teacher distillation, was repeated a second time to compress the 2B intermediate model down to the final 1.14B embedding model. Final training used a progressive two-stage context-scaling schedule:同样的序列（ModelOpt 结构化剪枝后进行 8B 教师蒸馏）被重复第二次，将 2B 中间模型压缩至最终的 1.14B 嵌入模型。最终训练采用了渐进式两阶段上下文缩放计划：

• Stage 1: Focused on broad multilingual alignment at 1024-token context length to reconstruct the core retrieval behavior of the parent model. 第一阶段：专注于 1024 Token 上下文长度下的广泛多语言对齐，以重建父模型的核心检索行为。
• Stage 2: Expanded context length to 4096 tokens and added long-context synthetic and reasoning datasets, helping the 1B model retain discriminative recall across longer inputs.第二阶段：将上下文长度扩展至 4096 Token，并增加了长上下文合成和推理数据集，帮助 1B 模型在更长的输入中保持辨别性召回能力。

The following table summarizes the core technical specifications and deployment targets for the Nemotron 3 Embed models:下表总结了 Nemotron 3 Embed 模型的核心技术规格和部署目标：

Table 2. Architectural specifications and core inference configurations for the Nemotron 3 Embed models.表 2. Nemotron 3 Embed 模型的架构规格与核心推理配置。

Enterprise Partner Evaluations 企业合作伙伴评估

Enterprise ISVs, AI-native companies, and memory providers are already evaluating Nemotron 3 Embed across agentic retrieval, agent memory, code retrieval, and production inference workflows. 企业级 ISV、AI 原生公司和内存提供商已经在智能体检索、智能体记忆、代码检索和生产推理工作流中评估 Nemotron 3 Embed。

• "Context is the key to agentic accuracy. Our Context Intelligence Graph uses embeddings and semantic similarity to deliver the most relevant enterprise context to agents like EnterpriseClaw, which we launched with NVIDIA in May. Early results from NVIDIA’s new Nemotron 3 Embed models are promising, particularly for question answering, where they show improvements over our current model. We’re excited about their potential to further improve the accuracy and reliability of our enterprise agents." - Adi Kuruganti, Chief AI and Development Officer at Automation Anywhere “上下文是智能体准确性的关键。我们的上下文智能图（Context Intelligence Graph）使用嵌入和语义相似度，为智能体（如我们五月与 NVIDIA 共同发布的 EnterpriseClaw）提供最相关的企业上下文。NVIDIA 新推出的 Nemotron 3 Embed 模型早期表现令人期待，特别是在问答方面，它们比我们目前的模型有所改进。我们对其进一步提升企业智能体准确性和可靠性的潜力感到兴奋。” - Adi Kuruganti，Automation Anywhere 首席 AI 与开发官
• "Our initial evaluation of the NVIDIA Nemotron 3 Embed models shows strong retrieval performance for our agentic retrieval use cases. The availability of both 1B and 8B variants gives teams the flexibility to balance quality, latency, and deployment requirements across different environments. We're excited to continue evaluating the models and exploring how they can support high-performance retrieval for production AI applications." - Mani Gill, Senior Vice President of Product Management at Boomi “我们对 NVIDIA Nemotron 3 Embed 模型的初步评估显示，其在我们的智能体检索用例中表现出强大的检索性能。1B 和 8B 变体的可用性，使团队能够灵活地平衡不同环境下的质量、延迟和部署需求。我们很高兴能继续评估这些模型，并探索它们如何支持生产级 AI 应用的高性能检索。” - Mani Gill，Boomi 产品管理高级副总裁
• IBM has seen promising early results evaluating the new NVIDIA Nemotron Embed model in a proof-of-concept built on watsonx.data. IBM 在基于 watsonx.data 构建的概念验证中评估了新的 NVIDIA Nemotron Embed 模型，并取得了令人期待的早期成果。
• "We ran Nemotron-3-Embed-1B inside our own stack against various models and it came out on top. For instance, it scored 80.38% against Qwen-3-0.6B’s 78.71% on LongMemEval for Retrieval@10. The open weights and fine-tuning recipes are what interest us most, since they let us adapt the model to memory text and serve it ourselves." - Rudraj Mehta, Product Manager at Mem0“我们在自己的栈中针对各种模型运行了 Nemotron-3-Embed-1B，结果它名列前茅。例如，在 LongMemEval 的 Retrieval@10 指标上，它得分为 80.38%，而 Qwen-3-0.6B 为 78.71%。开放权重和微调方案是我们最感兴趣的，因为它们让我们能够将模型适配到记忆文本并自行服务。” - Rudraj Mehta，Mem0 产品经理
• Palantir is collaborating with NVIDIA to evaluate the deployment of Nemotron 3 Embed for end customers conducting edge retrieval workloads, building on the prior work making NVIDIA embedding models available in AIP's Model Catalog. Palantir 正在与 NVIDIA 合作，评估 Nemotron 3 Embed 在终端客户执行边缘检索工作负载中的部署，并在此前将 NVIDIA 嵌入模型引入 AIP 模型目录的工作基础上进一步推进。
• ServiceNow is evaluating NVIDIA Nemotron Embed to retrieve through their documentation, with a continued interest in in-domain fine-tuning to improve their retrieval accuracy. ServiceNow 正在评估 NVIDIA Nemotron Embed 以便检索其文档，并持续关注领域内微调以提高其检索准确性。
• turbopuffer is bringing Nemotron 3 Embed to its native embeddings service, enabling developers to use the models within its semantic search engine. turbopuffer 正在将 Nemotron 3 Embed 引入其原生嵌入服务，使开发者能够在语义搜索引擎中使用这些模型。
• "Swapping in NVIDIA Nemotron 3 Embed model within our re-ranking stack delivered a significant leap in performance, letting us select query-relevant snippets from web pages with far greater accuracy than any model we'd used before." - Rahul Mohan, Senior AI Engineer, You.com “在我们的重排序栈中替换 NVIDIA Nemotron 3 Embed 模型带来了显著的性能飞跃，使我们能够比以往使用的任何模型更准确地从网页中选择与查询相关的片段。” - Rahul Mohan，You.com 高级 AI 工程师
• Zep evaluated NVIDIA Nemotron 3 Embed 1B for agent memory and context retrieval. In early internal benchmarks against several models used by Zep, including some much larger than Nemotron 3 Embed 1B, the model ranked first across all memory retrieval tasks. Zep believes the availability of an FP4 checkpoint also makes the model very storage and latency competitive for their scale. Zep 评估了 NVIDIA Nemotron 3 Embed 1B 用于智能体记忆和上下文检索。在针对 Zep 使用的多个模型（包括一些远大于 Nemotron 3 Embed 1B 的模型）的早期内部基准测试中，该模型在所有记忆检索任务中均排名第一。Zep 认为 FP4 检查点的可用性也使该模型在其规模下具备极强的存储和延迟竞争力。
• Zoom is evaluating NVIDIA Nemotron 3 Embed for the retrieval layer behind the enterprise agentic search that powers Zoom's contextual layer, which helps agents retrieve relevant context across meetings and workplace knowledge sources, including docs, chat, tickets, and other internal systems.Zoom 正在评估 NVIDIA Nemotron 3 Embed，用于支持 Zoom 上下文层背后的企业智能体搜索的检索层。该层帮助智能体跨会议和工作场所知识源（包括文档、聊天、工单和其他内部系统）检索相关上下文。

Getting Started 入门指南

NVIDIA Nemotron 3 Embed is released with open weights and open-source training recipes, giving organizations full control over how retrieval models are customized and deployed for production AI applications.NVIDIA Nemotron 3 Embed 发布时附带开放权重和开源训练方案，使组织能够完全控制如何为生产级 AI 应用定制和部署检索模型。

Developers can get started using the deployment option that best fits their workflow:开发者可以根据最适合其工作流的部署选项开始使用：

• Hugging Face - Access the model weights, model cards, and example code for SentenceTransformers, Transformers, and vLLM. Hugging Face - 获取模型权重、模型卡以及 SentenceTransformers、Transformers 和 vLLM 的示例代码。
• NVIDIA NIM - Access optimized microservice on build.nvidia.com for production-ready inference. NVIDIA NIM - 在 build.nvidia.com 访问优化后的微服务，进行生产就绪的推理。
• AI Cloud and Inference Partners - Deploy Nemotron 3 Embed through leading ecosystem partners - Baseten, Bitdeer AI, DeepInfra, Friendli AI, OpenRouter.AI 云与推理合作伙伴 - 通过领先的生态系统合作伙伴（Baseten、Bitdeer AI、DeepInfra、Friendli AI、OpenRouter）部署 Nemotron 3 Embed。

For workloads requiring domain adaptation or footprint reduction, we have also open-sourced NVIDIA NeMo AutoModel training recipes:对于需要领域适配或缩减模型规模的工作负载，我们还开源了 NVIDIA NeMo AutoModel 训练方案：

• Fine-tuning recipe - Adapt Nemotron 3 Embed to your enterprise corpus and retrieval tasks. 微调方案 - 将 Nemotron 3 Embed 适配到您的企业语料库和检索任务。
• Distillation recipe - Compress larger retrieval models while retaining ranking quality for production deployment.蒸馏方案 - 在保持生产部署排序质量的同时压缩大型检索模型。

For example, on the NV Docs evaluation, fine-tuning Nemotron-3-Embed-1B-BF16 improved NDCG@10 from 56.7% to 63.3% (+11.6%) and Recall@5 from 56.1% to 62.8% (+11.9%).例如，在 NV Docs 评估中，微调 Nemotron-3-Embed-1B-BF16 将 NDCG@10 从 56.7% 提高到 63.3%（+11.6%），将 Recall@5 从 56.1% 提高到 62.8%（+11.9%）。

Whether you are building enterprise search, production RAG, agent memory, code retrieval, or agentic AI systems, NVIDIA Nemotron 3 Embed provides flexible deployment options—from open-source models on Hugging Face to fully managed AI Cloud platforms and NVIDIA NIM microservices.无论您是在构建企业搜索、生产级 RAG、智能体记忆、代码检索还是智能体 AI 系统，NVIDIA Nemotron 3 Embed 都提供了灵活的部署选项——从 Hugging Face 上的开源模型到完全托管的 AI 云平台和 NVIDIA NIM 微服务。

Explore the models, deploy them on your preferred platform, and let us know what you are building.探索这些模型，在您首选的平台上部署它们，并让我们了解您的构建成果。