Intelligence
9
score
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) scores 9 (estimated) on the Artificial Analysis …
vs class avg: 25
NVIDIA•Open weights•April 7, 2025
Llama Nemotron Ultra
Intelligence, Performance & Price Analysis
Source:Artificial Analysis scrapeType:Open weights
Speed
50.9
output tokens/sec
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) generates output at 50. …
Latency
2.43s
TTFT
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) has a time to first token (TTFT) of 2. …
Input Price#56/93
$0.60
/ 1M tokens
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) costs $0.60 per 1M input tokens (somewhat higher than …
Output Price#52/93
$1.80
/ 1M tokens
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) costs $0.60 per 1M input tokens (somewhat higher than …
Decision Overview
Ethen's intelligence-driven routing assessment for this model
Overall verdict
Budget-friendly / task-specific model
Best for high-volume, simple, or domain-specific tasks where cost or speed matters more than deep reasoning.
Best for
- Simple Q&A
- High-throughput chat
- Classification
- Extraction
Weak or unsuitable
- Complex reasoning
- Creative writing
- Multi-step tasks
- Predictable high-volume throughput
Routing recommendation
1
Cheaper routes for predictable extraction, labeling, or summarization
2
Cache-backed reuse for repeated prompts
Cost pressure
Moderate pricing — $0.60 / M input, $1.80 / M output. Costs are manageable, but volume should still be reviewed.
Model-fit matrix
Use this model where capability pays for itself; route away where volume, latency, or budget dominate.
| Use case | Fit | Reason | Routing note |
|---|---|---|---|
| Complex reasoning & agentic workflows | Weak | Intelligence score 9 is better suited for straightforward tasks than multi-step reasoning. | Avoid routing complex agentic tasks to this model. |
| High-volume chat & customer-facing | Strong | Output speed 50.9 tokens/sec is adequate for chat. | Suitable for chat; monitor latency under concurrent load. |
| Latency-sensitive applications | Strong | TTFT 2.43s — adequate latency for most interactive use cases. | Suitable for real-time; test with your specific workload. |
| Cost-sensitive pipelines | Strong | Output pricing at $1.80 is reasonable for moderate volume. | Suitable for production; review costs as volume grows. |
Cost pressure
MediumPricing is moderate — input $0.60, output $1.80. Costs accumulate at volume but are manageable for valuable tasks.
Cheaper substitutes
- Cheaper routes for predictable extraction, labeling, or summarization
Route away when
- Predictable high-volume throughput
- Non-critical classification and extraction
Highlights
Key benchmark results at a glance.
Intelligence
Intelligence position
Artificial Analysis Intelligence Index · Higher is better · Evaluation results measured independently by Artificial Analysis
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) scores 9 (estimated) on the Artificial Analysis Intelligence Index, placing it at the lower end among other open weight models of similar size (median: 25).
Speed
Speed position
Output tokens per second · Higher is better · Evaluation results measured independently by Artificial Analysis
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) generates output at 50.9 tokens per second (based on the median across providers serving the model), which is at the lower end compared to other open weight models of similar size (median: 61.7 t/s).
Context, Cost & Pricing
Price/value position
Weighted average cost (USD) per Intelligence Index task · Lower is better · Evaluation results measured independently by Artificial Analysis
Pricing is moderate — input $0.60, output $1.80. Costs accumulate at volume but are manageable for valuable tasks.
Intelligence & Quality
Benchmark scores and quality indices measuring model capability.
AA-Omniscience Index
AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct. · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Bars intentionally show many models at once so the page feels like a benchmark report rather than a dashboard.
Artificial Analysis Intelligence Index by Open Weights / Proprietary
Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Artificial Analysis Intelligence Index
Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Artificial Analysis Openness Index: Score
Openness Index assesses model openness on a 0 to 100 normalized scale (higher is more open) · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Bars intentionally show many models at once so the page feels like a benchmark report rather than a dashboard.
Intelligence
Artificial Analysis Intelligence Index · Higher is better · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Speed & Throughput
Output tokens per second and generation throughput.
Output Speed
Output tokens per second · Higher is better · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Speed should be read together with latency and cost; fast output does not always mean better user-perceived performance.
Speed
Output tokens per second · Higher is better · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Speed should be read together with latency and cost; fast output does not always mean better user-perceived performance.
Latency & Response Time
Time to first token and end-to-end response latency.
End-to-End Response Time
Seconds to output 500 tokens, including reasoning model 'thinking' time · Lower is better · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Bars intentionally show many models at once so the page feels like a benchmark report rather than a dashboard.
Latency: Time To First Answer Token
Seconds to first answer token received · Accounts for reasoning model 'thinking' time · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Latency includes thinking / reasoning time when applicable. Lower is better for interactive use cases.
Context, Cost & Pricing
Context window size, cost per task, and token pricing.
Context Window
Context window: tokens limit · Higher is better · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Larger context windows enable complex multi-turn tasks but may increase latency and cost for long inputs.
Cost per Intelligence Index Task
Weighted average cost (USD) per Artificial Analysis Intelligence Index task, segmented by token type. Lower is better · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.
Cost per Task
Weighted average cost (USD) per Intelligence Index task · Lower is better · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.
Cost to Run Artificial Analysis Intelligence Index
Cost (USD) to run all evaluations in the Artificial Analysis Intelligence Index · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.
Pricing: Cache Hit, Input, and Output
Price (USD per M Tokens) · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Bars intentionally show many models at once so the page feels like a benchmark report rather than a dashboard.
Time per Intelligence Index Task
Weighted average decode time (minutes) per task; excludes TTFT and overhead time · Lower is better · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Architecture & Scale
Model size, parameters, and architectural details.
Model Size: Total and Active Parameters
Comparison between total model parameters and parameters active during inference · Evaluation results measured independently by Artificial Analysis
Source and methodology are summarized below
Chart source and provenance are listed in Methodology & sources below.
Larger models generally perform better but cost more to serve. Active parameter count matters for inference efficiency.
Specifications
Technical Specifications
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) scores 9 (estimated) on the Artificial Analysis Intelligence Index, placing it at the lower end among other open weight models of similar size (median: 25). Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) generates output at 50.9 tokens per second (based on the median across providers serving the model), which is at the lower end compared to other open weight models of similar size (median: 61.7 t/s). Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) costs $0.60 per 1M input tokens (somewhat higher than average, median: $0.59) and $1.80 per 1M output tokens (better than average, median: $2.20), based on the median across providers serving the model.
Llama Nemotron Ultra
- Model type
- Open weights
- Reasoning
- Yes
- Input modalities
- Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) supports text input.
- Output modalities
- Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) supports text output.
- Context window
- 130k tokens
- Open weights / source
- Yes, Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
- Parameters
- Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) has 253 billion parameters.
- License
- Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) is released under the NVIDIA Open Model License Agreement license. This license allows commercial use.
- API availability
- Yes, Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) is available via API through 1 provider.
Methodology & Provenance
This page is rendered from the normalized profile and page JSON for Llama Nemotron Ultra.
Benchmark values are preserved as normalized; only layout, disclosure ordering, and typography are adjusted for readability.
Source file: md/scrape/artificial_analysis_model_pages/artificialanalysis_ai__models__llama-3-1-nemotron-ultra-253b-v1-reasoning__eaf78a1843.html · Canonical URL: https://artificialanalysis.ai/models/llama-3-1-nemotron-ultra-253b-v1-reasoning
Normalized at: 2026-07-08T03:24:56+00:00 Technical specs, comparison summary, and FAQs are passed through from the normalized data without rewriting.
Ranks shown on summary cards come directly from the normalized profile. No rank or class display is synthesized in the renderer.
Visible FAQ answers and FAQ JSON-LD are emitted from the same normalized FAQ array, so structured data stays aligned with the rendered accordion.
Time per Intelligence Index Task: https://artificialanalysis.ai/models/llama-3-1-nemotron-ultra-253b-v1-reasoning
Frequently Asked Questions
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) was released on April 7, 2025.
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) was created by NVIDIA.
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) scores 9 (estimated) on the Artificial Analysis Intelligence Index, placing it at the lower end among other open weight models of similar size (median: 25).