Nemotron 3 Ultra
NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...
Updated 1 h ago · first seen 11 Sept 2026
model_01M294WVXG0ZTDNK5X5S7V5438
- Context
- 262.1K tokens
- T2 · 3 h ago
- Released
- 4 Jun 2026
- T2 · 3 h ago
Specification
- Release date
- 4 Jun 2026
Source:OpenRouter public model & pricing listingT2observed 3 h agomedium
- Openness
- open-weights
Source:Artificial AnalysisT2observed 1 h agomedium
- Context window
- 262.1K tokens
Source:OpenRouter public model & pricing listingT2observed 3 h agomedium
- Max output
- 32.8K tokens
Source:OpenRouter public model & pricing listingT2observed 3 h agomedium
- Modalities
- text
Source:OpenRouter public model & pricing listingT2observed 3 h agomedium
- Input modalities
- text
Source:OpenRouter public model & pricing listingT2observed 3 h agomedium
- Output modalities
- text
Source:OpenRouter public model & pricing listingT2observed 3 h agomedium
- Hugging Face repo
- nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16
Source:OpenRouter public model & pricing listingT2observed 3 h agomedium
- Livebench hf link
- https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4
Source:LiveBenchT2observed 1 h agomedium
- Aa median output tokens per second
- 160.4
Source:Artificial AnalysisT2observed 1 h agomedium
- Openrouter id
- nvidia/nemotron-3-ultra-550b-a55b
Source:OpenRouter public model & pricing listingT2observed 3 h agomedium
- Supported parameters
- frequency_penalty, include_reasoning, logit_bias, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_p
Source:OpenRouter public model & pricing listingT2observed 3 h agomedium
Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →
Provenance
Attributed facts
16
Source tiers
T216
Freshest observation
1 h ago
Conflicts
None
Modalities
- Modalities
- text
- Input
- text
- Output
- text
Capabilities
Tool calling
Yes
OpenRouter public model & pricing listing · T2
Structured output
Yes
OpenRouter public model & pricing listing · T2
Reasoning
Yes
OpenRouter public model & pricing listing · T2
Vision
Unavailable
Audio
Unavailable
Fine-tuning available
Unavailable
- Context window
- 262.1K tokens
Source:OpenRouter public model & pricing listingT2observed 3 h agomedium
- Max output
- 32.8K tokens
Source:OpenRouter public model & pricing listingT2observed 3 h agomedium
| Benchmark | Score | Metric | Config | Evaluated | Source |
|---|---|---|---|---|---|
| τ²-bench | 83.33% | pass^1 | aa_slug=nvidia-nemotron-3-ultra-550b-a55b · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
| Artificial Analysis Intelligence Index | 23.41 | index | aa_slug=nvidia-nemotron-3-ultra-550b-a55b · version=4.3 · estimated=false | — | artificialanalysis.aiT2 |
| GPQA | 86.67% | accuracy | aa_slug=nvidia-nemotron-3-ultra-550b-a55b · variant=GPQA Diamond · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
| Humanity's Last Exam | 28.41% | accuracy | aa_slug=nvidia-nemotron-3-ultra-550b-a55b · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
| IFBench | 81.36% | accuracy | aa_slug=nvidia-nemotron-3-ultra-550b-a55b · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
| LiveBench | 54.46% | category:Data Analysis | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=nemotron-3-ultra-550b-a55b | 25 Jun 2026 | livebench.aiT2 |
| LiveBench | 67.36% | global_average | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=nemotron-3-ultra-550b-a55b | 25 Jun 2026 | livebench.aiT2 |
| LiveBench | 74.7% | category:Reasoning | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=nemotron-3-ultra-550b-a55b | 25 Jun 2026 | livebench.aiT2 |
| LiveBench | 88.66% | category:Mathematics | release=2026-06-25 · subtasks=["AMPS_Hard","integrals_with_game","math_comp","olympiad"] · livebench_model_id=nemotron-3-ultra-550b-a55b | 25 Jun 2026 | livebench.aiT2 |
| LiveBench | 73.45% | category:IF | release=2026-06-25 · subtasks=["paraphrase","simplify","story_generation","summarize"] · livebench_model_id=nemotron-3-ultra-550b-a55b | 25 Jun 2026 | livebench.aiT2 |
| LiveBench | 70.81% | category:Language | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=nemotron-3-ultra-550b-a55b | 25 Jun 2026 | livebench.aiT2 |
| LiveBench | 38.74% | category:Agentic Coding | release=2026-06-25 · subtasks=["javascript","typescript","python"] · livebench_model_id=nemotron-3-ultra-550b-a55b | 25 Jun 2026 | livebench.aiT2 |
| LiveBench | 70.7% | category:Coding | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=nemotron-3-ultra-550b-a55b | 25 Jun 2026 | livebench.aiT2 |
| SciCode | 40.28% | accuracy | aa_slug=nvidia-nemotron-3-ultra-550b-a55b · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
| Terminal-Bench | 36.36% | accuracy | aa_slug=nvidia-nemotron-3-ultra-550b-a55b · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
| Terminal-Bench | 0.51% | accuracy | aa_slug=nvidia-nemotron-3-ultra-550b-a55b · variant=v4.0 · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
| Terminal-Bench | 53.93% | accuracy | aa_slug=nvidia-nemotron-3-ultra-550b-a55b · variant=v2.1 · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
Scores are reported as published, with their evaluation configuration (harness, prompting, judge). Results with different configs are not directly comparable — see methodology.
Current prices
| Provider | Input / 1M | Output / 1M | Cached in | Batch in / out | Context | Observed | Source |
|---|---|---|---|---|---|---|---|
| NVIDIA NIM / build.nvidia.comnvidia/nemotron-3-ultra-550b-a55b:free | $0 | $0 | — | — | 1M | 1 h ago | openrouter.aiT2 |
| NVIDIA NIM / build.nvidia.comnvidia/nemotron-3-ultra-550b-a55b | $0.625 | $3.13 | $0.188 | — | 256K | 1 h ago | openrouter.aiT2 |
USD per 1M tokens as published by each provider (USD). Rows are append-only: every change is kept in the history below.
Available through 1
Price history
Input price · USD / 1M tokens
Output price · USD / 1M tokens
- NVIDIA NIM / build.nvidia.com
No hardware estimate available
No lineage recorded
Papers 0
No papers linked yet.
Repositories 0
No repositories linked yet.
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history · Modalities
Modalitiesmodalities1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| text | → current | current | OpenRouter public model & pricing listingT2 | medium | deterministic |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
Nemotron 3 Ultra scores 73.445% on LiveBench
livebench_leaderboardNemotron 3 Ultra scores 70.811% on LiveBench
livebench_leaderboardNemotron 3 Ultra scores 54.457% on LiveBench
livebench_leaderboardNemotron 3 Ultra scores 88.663% on LiveBench
livebench_leaderboardNemotron 3 Ultra scores 38.737% on LiveBench
livebench_leaderboardNemotron 3 Ultra scores 70.698% on LiveBench
livebench_leaderboardNemotron 3 Ultra scores 74.697% on LiveBench
livebench_leaderboardNemotron 3 Ultra scores 67.358% on LiveBench
livebench_leaderboardNemotron 3 Ultra scores 36.36% on Terminal-Bench
artificial_analysisNemotron 3 Ultra scores 53.93% on Terminal-Bench
artificial_analysisNemotron 3 Ultra scores 0.51% on Terminal-Bench
artificial_analysisNemotron 3 Ultra scores 83.33% on τ²-bench
artificial_analysisNemotron 3 Ultra scores 81.36% on IFBench
artificial_analysisNemotron 3 Ultra scores 40.28% on SciCode
artificial_analysisNemotron 3 Ultra scores 28.41% on Humanity's Last Exam
artificial_analysisNemotron 3 Ultra scores 86.67% on GPQA
artificial_analysisNemotron 3 Ultra scores 23.41 on Artificial Analysis Intelligence Index
artificial_analysisNVIDIA NIM / build.nvidia.com lists Nemotron 3 Ultra at $0 in / $0 out per 1M tokens
openrouterNVIDIA NIM / build.nvidia.com lists Nemotron 3 Ultra at $0.625 in / $3.125 out per 1M tokens
openrouter
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| LiveBench | livebench.ai/table_2026_06_25.csv | leaderboard | T2· Quality secondary | 1 h ago | 1 |
| Artificial Analysis | artificialanalysis.ai/leaderboards/models | leaderboard | T2· Quality secondary | 1 h ago | 1 |
| OpenRouter public model & pricing listing | openrouter.ai/api/v1/models | catalogue | T2· Quality secondary | 1 h ago | 2 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.