NVIDIA Nemotron 3 Ultra Powers Faster, More Efficient Reasoning for Long-Running Agents

NVIDIA released the Nemotron 3 Ultra, a 550-billion-parameter open model designed to accelerate reasoning and reduce costs for long-running AI agents. The model achieves up to five times higher throughput than comparable systems while lowering token consumption and operational expenses by as much as 30%. The release addresses the growing computational demands of multi-agent workflows, where continuous tool calls and context accumulation drive up costs and risk goal drift.

Single-turn chatbots are evolving into long-running agents https://www.nvidia.com/en-us/glossary/ai-agents/ that can reason, maintain context, use tools, and run efficiently across many turns to complete complex workflows. However, these multi-agent https://www.nvidia.com/en-us/glossary/multi-agent-systems/ workflows cause token counts to grow quickly. Agents plan, call tools, invoke sub-agents, receive information, and then pass history, outputs, and reasoning https://www.nvidia.com/en-us/glossary/ai-reasoning/ steps back into the model continuously. As tasks run longer, this constant communication increases costs and the risk of goal drift. Developers can solve this using a system of models: frontier reasoning models https://www.nvidia.com/en-us/glossary/frontier-models/ for orchestration and complex planning, and efficient models for high-volume execution, validation, and tool calling. NVIDIA is releasing NVIDIA Nemotron 3 Ultra, an open model built to help long-running agents complete tasks faster while lowering cost. Nemotron 3 Ultra for agent orchestration Nemotron 3 Ultra is a 550B-parameter Mixture-of-Experts model with 55B active parameters, built for frontier reasoning and orchestration in agentic systems. Within any agent workflow, most calls are routine, but a critical subset demands deeper reasoning. Nemotron 3 Ultra is built to handle these hard calls: sustaining architectural decisions across coding sessions, synthesizing contradictory evidence across hundreds of research sources, or verifying chip designs across thousands of constraints. | Nemotron 3 Ultra 550B | GLM 5.1 744B | Kimi K2.6 1T | Qwen3.5 397B | | | Agent Productivity PinchBench | 91% | 84% | 91% | 89% | | Long-horizon Planning EnterpriseOps-Gym | 33% | 40% | 29% | 30% | | Coding Terminal-Bench 2.0 | 54% | 64% | 67% | 53% | | Instruction Following IFBench | 82% | 77% | 74% | 78% | | Knowledge Work GDPVal-AA | 1,448 | 1,594 | 1,508 | 1,192 | | Professional Work Tasks ProfBench Search | 56% | 46% | 56% | 53% | | Long Context Ruler @1M | 95% | N/A max 256K | N/A max 256K | 90% | Table 1. Nemotron 3 Ultra delivers frontier accuracy in a smaller model Nemotron 3 Ultra is also fast. It achieves 5x higher throughput compared to other open models in its class, enabling long-running agents to complete tasks faster and more efficiently. Nemotron 3 Ultra is also built for efficiency. In experiments on the SWE-bench and Terminal bench 2.0, it completed benchmarks using fewer total tokens and fewer tokens per turn than comparable models. This lowers the cost for agentic tasks by up to 30%. Breakthroughs powering Nemotron 3 Ultra To mitigate the typical efficiency-accuracy tradeoffs for high-capacity reasoning models, the Nemotron models introduce architectural innovations: Post-trained for agent harness Nemotron Ultra is post-trained to deliver consistent accuracy across top harnesses. The model is trained using the NVIDIA NeMo RL https://github.com/nvidia-nemo/rl and Gym https://github.com/NVIDIA-NeMo/gym open libraries with one of the largest suites of long-running, task-solving, tool-using datasets in the world. Ultra is optimized for agent-led open harnesses, not just single-turn chat, and is designed to work within workflows where agents plan, call tools, read observations, delegate to sub-agents, validate outputs, and recover from errors across many turns. Hybrid Mamba transformer Mamba layers improve sequence efficiency for long-context workloads, while Transformer layers preserve precise recall when agents need to retrieve specific facts from large context windows. NVFP4 precision The same NVFP4 checkpoint runs on NVIDIA Hopper, NVIDIA Blackwell, and Ampere GPUs. Developers can use one checkpoint across all NVIDIA GPU architectures thanks to specialized NVFP4 quantization kernels. NVFP4 also delivers up to 5x higher throughput per GPU at the same interactivity compared to BF16 on Blackwell. LatentMoE LatentMoE supports more efficient expert routing, enabling the model to handle workflows spanning reasoning, code generation, tool calls, and domain-specific logic. Multi-token prediction Multi-token prediction MTP helps reduce generation time by predicting multiple future tokens in a single forward pass, improving throughput for long outputs and multi-turn workflows. Nemotron 3 Ultra adds Multi-Teacher On-Policy Distillation Multi-Teacher On-Policy Distillation MOPD is a training method in which Ultra learns from multiple specialized teacher models while generating its own attempts during training. More than 10 specialized teacher models are trained, each with its own domain-specific training pipeline. Each teacher scores the model in its area of expertise, helping Ultra improve reasoning across domains more efficiently. During MOPD, the student model generates rollouts across domains and receives dense reward signals from the corresponding teacher models. To maximize efficiency, MOPD runs asynchronously, with student rollout generation, teacher scoring, and student optimization fully pipelined. MOPD is also iterative. After producing an MOPD-trained checkpoint, new rounds of teacher training are initialized from the updated student model, and the improvements are merged into the next MOPD stage. This co-evolution between students and teachers enables continuous capability improvement and progressively stronger specialization across domains. Users can try MOPD recipes https://github.com/NVIDIA-NeMo/RL/blob/ultra-v3/docs/guides/nemotron-3-ultra.md through NeMo-RL, the library that trained the Ultra model. Training data for stronger agent reasoning As with all Nemotron open model launches, much of the training data pipeline is released as permissively as possible. For partners in enterprise and sovereign AI development, training data transparency and provenance matter as much as capability. Domain-specific pre-training data Building on a 10T token pre-training foundation, Nemotron 3 Ultra adds 212B new tokens targeting three high-value domain gaps: 4B tokens of synthetic legal data, increasing the proxy LegalBench average from 64.6% to 74.7% 35B tokens of synthesized Wiki-based data, boosting proxy SimpleQA from 40.2% to 50.2% 173B refreshed GitHub tokens through Sept. 30, 2025 Post-training data and RL environments This launch is also releasing 10M new SFT samples, 1M new RL tasks across multiple domains, and 15 net-new RL environments, bringing the cumulative Nemotron open data totals to 50M SFT samples, 2M RL tasks, and 55 RL environments. The result is SWEBench Verified scores between 65% and 70.4% across Pi, OpenHands, Hermes, OpenCode, and Mini SWE Agent—consistent performance regardless of which framework you deploy. Finetune for your domain Nemotron 3 Ultra can be fine-tuned using LoRA, SFT, and reinforcement learning using the NVIDIA NeMo libraries https://github.com/NVIDIA-NeMo . Developers can get started with the following recipes. Nemotron 3 Ultra Recipes: - SFT LoRA: NeMo Automodel H100 Recipe https://github.com/NVIDIA-NeMo/Automodel/blob/main/examples/llm finetune/nemotron/nemotron ultra v3 hellaswag peft.yaml , GB200 Recipe https://github.com/NVIDIA-NeMo/Automodel/blob/main/examples/llm finetune/nemotron/nemotron ultra v3 hellaswag peft gb200.yaml - Full SFT: NeMo Megatron Bridge Recipes https://github.com/NVIDIA-NeMo/Megatron-Bridge/tree/nemotron 3 ultra/examples/models/nemotron/nemotron 3/ultra - Reinforcement Learning: NeMo RL GRPO recipe https://github.com/NVIDIA-NeMo/RL/blob/ultra-v3/docs/guides/nemotron-3-ultra.md , GRPO LoRA recipe https://github.com/NVIDIA-NeMo/Nemotron/tree/main/usage-cookbook/Nemotron-3-Ultra , MOPD recipe https://github.com/NVIDIA-NeMo/RL/blob/ultra-v3/docs/guides/nemotron-3-ultra.md - Deployment: Dynamo Recipe https://github.com/ai-dynamo/dynamo/tree/main/recipes/nemotron-3-ultra See it in action This walkthrough shows how to spin up and run an autoresearch flow using Hermes Agent powered by Nemotron 3 Ultra on build.nvidia.com https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b . Run agents more safely with NVIDIA NemoClaw and NVIDIA OpenShell Nemotron models integrate with leading open agent frameworks. To build a secure, always-on agentic system, it is important to understand the reference stack: Hermes Agent and OpenClaw: These are popular agent harnesses that provide the orchestration loops, memory, and tools for multi-turn workflows. Hermes Agent is now officially available and fully supported for use with Nemotron. NVIDIA OpenShell: Available now in early preview, OpenShell is the secure runtime environment part of the NVIDIA Agent Toolkit where autonomous agents and their generated code execute. NVIDIA NemoClaw: This is the open-source blueprint that ties the environment together. With a single command, NemoClaw installs the OpenShell runtime—providing a secure environment for running autonomous agents like Hermes Agent more safely alongside open-source models like Nemotron. Build safer and voice-enabled agents Two new Nemotron models are also launching: Nemotron 3.5 Content Safety For teams building safer enterprise AI, Nemotron 3.5 Content Safety https://huggingface.co/nvidia/Nemotron-3.5-Content-Safety is an open, efficient 4B guardrail model for classifying unsafe, disallowed, or policy-violating content across text, images, and combined inputs. Covering 23 safety categories and 12 languages, it can be used as an inference-time guardrail, as a judge for LLM safety testing and evaluation, or with the accompanying training dataset to post-train models for safer behavior. Custom policy support and reasoning trails help enterprises adapt safety decisions to domain-specific rules, audit classifications, and deploy safety controls across global AI workflows. Read the Hugging Face post https://huggingface.co/blog/nvidia/nemotron-3.5-content-safety to learn more. Nemotron 3.5 ASR For voice-native agents, Nemotron 3.5 ASR https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-ASR-Streaming-Multilingual-0.6b uses the same cache-aware streaming architecture as its English predecessor, Nemotron 3 ASR, to process audio deltas instantly. Eliminating redundant buffered compute ensures sub-100 ms latency for natural, real-time voice orchestration for your agentic swarms. The English model has seen strong developer adoption, including powering the voice input feature in Microsoft GitHub Copilot CLI https://aka.ms/FL BUILD 2026 , used by more than 20M developers. An independent benchmark of 50+ on-device ASR https://arxiv.org/html/2604.14493v2 configurations identified Nemotron 3 ASR as the strongest candidate for real-time English streaming on resource-constrained hardware. Now, that same architecture goes multilingual with support for 40+ languages in a single checkpoint. Updated open licensing for broader adoption Nemotron model releases are moving to OpenMDW-1.1, the Linux Foundation’s permissive license purpose-built for open AI model distributions. OpenMDW is designed to cover the full set of model materials, including architecture, parameters, documentation, software, and other related artifacts, under a single framework. This gives developers and enterprises clearer terms for using, modifying, redistributing, and deploying Nemotron models, while reducing the licensing ambiguity that can slow evaluation and adoption of open models. Start building today Nemotron 3 Ultra is fully open—including weights, data, and recipes—so developers can adapt the models to domain-specific workflows and deploy them anywhere. It is available across leading inference platforms and packaged as an NVIDIA NIM microservice, it can run anywhere. Try it on Perplexity https://www.perplexity.ai/ with a Pro subscription or through API, OpenRouter https://openrouter.ai/nvidia/nemotron-3-ultra-550b-a55b , Anaconda https://www.anaconda.com/blog/nvidia-nemotron-3-ultra-available-anaconda , or build.nvidia.com https://build.nvidia.com/nvidia/nemotron-3-ultra-550b-a55b .Download the weights from Hugging Face https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4 , launch an optimized instance through NVIDIA NIM, or start with the cookbooks to get running in minutes Download the weights from Hugging Face https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4 , launch an optimized instance through NVIDIA NIM, or start with the cookbooks to get running in minutes. Nemotron 3 Ultra is also available through AWS JumpStart, Amazon EKS, Baseten, Bitdeer AI, CoreWeave, Crusoe, DeepInfra, Dell Enterprise Hub, DigitalOcean, Eigen AI, fal ASR , Fireworks AI, FriendliAI, GMI Cloud, Google Cloud, Lightning AI, Microsoft Foundry, Modal, Nebius Token Factory, Prime Intellect, Simplismart, Together AI along with ASR , and Vultr. Check out the GitHub repository https://github.com/NVIDIA-NeMo/Nemotron/tree/main/usage-cookbook/Nemotron-3-Ultra/ for getting-started instructions for agent harness, including BlackBox AI https://www.blackbox.ai/blog/nemotron-on-blackbox , Cline https://docs.cline.bot/api/models , CrewAI https://docs.crewai.com/en/concepts/llms nvidia-nemotron , Factory AI https://docs.factory.ai/cli/user-guides/choosing-your-model , Hermes Agent https://unsloth.ai/docs/models/nemotron-3-ultra , Kilo Code https://kilo.ai/docs/gateway/models-and-providers , LangChain Deep Agents https://docs.langchain.com/oss/python/deepagents/models , OpenClaw, OpenCode https://opencode.ai/docs , OpenHands https://docs.openhands.dev/openhands/usage/llms/openrouter , and Pi https://pi.dev/models?provider=openrouter&name=nemotron+3+ultra . For the full technical details, read the Nemotron 3 Ultra technical report https://research.nvidia.com/labs/nemotron/files/NVIDIA-Nemotron-3-Ultra-Technical-Report.pdf . Stay up to date on NVIDIA Nemotron by subscribing to NVIDIA news and following NVIDIA AI on LinkedIn , X , Discord , and YouTube . Visit the Nemotron developer page for resources to get started. Explore open Nemotron models and datasets on Hugging Face and Blueprints on build.nvidia.com . Engage with Nemotron livestreams , tutorials , and the developer community on the NVIDIA forum and Discord .