NVIDIAEco TierMoE SparseZero Data Retention

Nemotron 3 Ultra

Name: ARMES
Availability: InStock
Author: ARMES

Run Nemotron 3 Ultra

Included on ECO plan

Context Window

Input Token Price

$0.60

per 1,000,000 tokens

Output Token Price

$3.60

per 1,000,000 tokens

Speed Rating

Fast

Fast (~300+ t/s; 5.9× higher throughput than comparable open models)

Architectural Profile & Capabilities

SWE-Bench Verified: 70.7–71.9%, PinchBench: 90.0%, RULER @1M: 94.7%, LiveCodeBench v6: 89.0%, IOI 2025: 570, GPQA (no tools): 87.0%, IFBench: 81.7%, IMOAnswerBench: 88.6–92.3%, Terminal-Bench 2.1: 56.4%, Artificial Analysis Intelligence Index: 48 (highest US open model) | NVIDIA's most capable model (released Jun 4, 2026). Same hybrid Mamba-Transformer LatentMoE architecture as Super, scaled to 550B/55B. **Frontier reasoning and orchestration for long-running agents at extreme throughput.** 5.9× faster inference than GLM-5.1; 30% lower cost-to-task-completion than comparable models. Strong at: agent orchestration, coding agents, deep research, complex enterprise tasks, multi-step reasoning, long-context analysis (RULER @1M: 94.7%), competitive math (IOI, IMOAnswerBench), and multilingual work (11 languages, 43 programming languages). Pre-trained on 20T tokens. OpenMDW 1.1 license (open weights). Configurable reasoning mode (on/off). **Caveat: Terminal-Bench 2.1 at 56.4% trails frontier proprietary models; best suited for high-volume agentic pipelines where throughput matters more than per-task ceiling.**

Architecture Type: Hybrid Mamba-Transformer

Sparse Mixture-of-Experts routing: activates only a subset of parameter experts per token, delivering frontier-level intelligence with exceptional efficiency.

Recommended Workloads & Primary Use Cases

Everyday reasoning

Drafting

Fast problem solving

ARMES Zero Data Retention

Calls to Nemotron 3 Ultra are routed through strict Zero Data Retention inference channels. Prompts and outputs are never stored, indexed, or monitored by NVIDIA.

View Privacy Architecture

More Models from NVIDIA

Compare other engines available in this laboratory.

View Roster (3)

Nemotron 3 SuperFree Tier

GPQA Diamond: 79.23% (82.70% w/ tools), HMMT Feb 2025: 93.67% (94.73% w/ tools), AIME 2025: 90.21%, MMLU-Pro: 83.73%, LiveCodeBench v5: 81.19%, SWE-Bench Verified: 60.47%, SWE-Bench Multilingual: 45.78%, RULER @1M: 91.75% (vs GPT-OSS-120B's 22.30), RULER @256K: 96.30%, AA Intelligence Index: 36 | Agentic reasoning, tool calling, long-context workflows, instruction following, multi-agent orchestration.

1M ctx$0.30/M

Nemotron 3.5 LightningFree Tier

SWE-Bench Verified: 51.56% (BF16) / 52.80% (NVFP4), PinchBench: 85.37%, GPQA Diamond: 75.44%, MMLU Pro: 81.94%, Terminal-Bench 2.1: 24.58%, HLE: 11.72%, IFBench (loose): 71.88%, BrowseComp: 36.97%, AA Intelligence Index: 23.6, AA Coding Index: 26.8, AA Agentic Index: 13.8, AA Non-Hallucination Rate: 62.4% | NVIDIA's highest-efficiency model, purpose-built for the execution layer of always-on agents (released Aug 11, 2026).

1M ctx$0.10/M

Open App

Back to NVIDIA models roster

NVIDIAEco TierMoE SparseZero Data Retention

Nemotron 3 Ultra

Run Nemotron 3 Ultra

Included on ECO plan

Context Window

Input Token Price

$0.60

per 1,000,000 tokens

Output Token Price

$3.60

per 1,000,000 tokens

Speed Rating

Fast

Fast (~300+ t/s; 5.9× higher throughput than comparable open models)

Architectural Profile & Capabilities

Architecture Type: Hybrid Mamba-Transformer

Sparse Mixture-of-Experts routing: activates only a subset of parameter experts per token, delivering frontier-level intelligence with exceptional efficiency.

Recommended Workloads & Primary Use Cases

Everyday reasoning

Drafting

Fast problem solving

ARMES Zero Data Retention

Calls to Nemotron 3 Ultra are routed through strict Zero Data Retention inference channels. Prompts and outputs are never stored, indexed, or monitored by NVIDIA.

View Privacy Architecture

More Models from NVIDIA

Compare other engines available in this laboratory.

View Roster (3)

Nemotron 3 SuperFree Tier

1M ctx$0.30/M

Nemotron 3.5 LightningFree Tier

1M ctx$0.10/M