nvidia · Open weights

Nvidia Nemotron Super 49B

Added Aug 8, 2025Open weights

Llama-3.3-Nemotron-Super-49B-v1 is a model which offers a great tradeoff between model accuracy and efficiency. Efficiency (throughput) directly translates to savings. Using a novel Neural Architecture Search (NAS) approach, we greatly reduce the model's memory footprint, enabling larger workloads, as well as fitting the model on a single GPU at high workloads (H200). This NAS approach enables the selection of a desired point in the accuracy-efficiency tradeoff. For more information on the NAS approach, please refer to this paper.

Input: Text
Output: Text
Discovery tier: Open weights

Pricing

USD per 1M tokens

Input: $0.15
Output: $0.15

Reference prices as published by the model provider. API example

Specifications

Context window: 128K
Max output: 16.4K

Side by side

Nvidia Nemotron Super 49B compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Limits
	Nvidia Nemotron Super 49BnvidiaThis model	Nvidia Nemotron 3.5 Lightning ThinkingnvidiaRemove
Provider	nvidia	nvidia
Discovery tier	Open weights	Open weights
Added	Aug 8, 2025	Aug 12, 2026
Context window	128K	1MBest
Max output	16.38K	65.54KBest
Average output	—	6.1K tokens
Pricing, USD per 1M tokens
Input	$0.15	$0.05Best
Output	$0.15Best	$0.20
Cache read	—	$0.01
Benchmark indexes
Intelligence Index	7.3	12.9Best
Coding Index	—	26.8
Math Index	7.7	—
Capabilities
Reasoning	No	Yes
Tools	No	Yes
Parallel tools	No	Yes
Reasoning effort	—	none, high

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

Nvidia Nemotron Super 49B
Nvidia Nemotron 3.5 Lightning Thinking

Intelligence Index

Nvidia Nemotron Super 49B7.3
Nvidia Nemotron 3.5 Lightning Thinking12.9

Nvidia Nemotron Super 49B: Better than 10% of 326 scored models

3.8Catalog models by score55.1

Indexes

Coding Index
Composite of the coding evaluations
- Nvidia Nemotron Super 49B: —
- Nvidia Nemotron 3.5 Lightning Thinking: 26.8
Math Index
Composite of the math evaluations
- Nvidia Nemotron Super 49B: 7.7
- Nvidia Nemotron 3.5 Lightning Thinking: —

Agentic work

AutomationBench-AA
Workflow automation, penalized for guardrail violations
- Nvidia Nemotron Super 49B: —
- Nvidia Nemotron 3.5 Lightning Thinking: 0.8%
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- Nvidia Nemotron Super 49B: —
- Nvidia Nemotron 3.5 Lightning Thinking: 0.0%
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- Nvidia Nemotron Super 49B: —
- Nvidia Nemotron 3.5 Lightning Thinking: 506 Elo
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- Nvidia Nemotron Super 49B: —
- Nvidia Nemotron 3.5 Lightning Thinking: 638 Elo

Document reasoning

GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- Nvidia Nemotron Super 49B: —
- Nvidia Nemotron 3.5 Lightning Thinking: 3.0%
AA-LCRv1.1
Reasoning across long documents
- Nvidia Nemotron Super 49B: 12.0%
- Nvidia Nemotron 3.5 Lightning Thinking: 60.3%Top
MLCR-AA
Long-context reasoning over medical records
- Nvidia Nemotron Super 49B: —
- Nvidia Nemotron 3.5 Lightning Thinking: 2.2%

Reasoning and knowledge

Humanity’s Last Exam
Expert-level questions across academic fields
- Nvidia Nemotron Super 49B: 3.8%
- Nvidia Nemotron 3.5 Lightning Thinking: 10.6%Top
GPQA Diamond
Graduate-level science questions
- Nvidia Nemotron Super 49B: 51.7%
- Nvidia Nemotron 3.5 Lightning Thinking: 74.3%Top
MMLU-Pro
Multi-subject knowledge and reasoning
- Nvidia Nemotron Super 49B: 69.8%
- Nvidia Nemotron 3.5 Lightning Thinking: —
IFBench
Following precise output instructions
- Nvidia Nemotron Super 49B: 39.5%
- Nvidia Nemotron 3.5 Lightning Thinking: —

Coding

Terminal-Benchv4.0
Practical software engineering in a terminal
- Nvidia Nemotron Super 49B: —
- Nvidia Nemotron 3.5 Lightning Thinking: 0.5%
SciCode
Scientific computing in Python
- Nvidia Nemotron Super 49B: —
- Nvidia Nemotron 3.5 Lightning Thinking: 32.1%
LiveCodeBench
Recent competitive programming problems
- Nvidia Nemotron Super 49B: 28.0%
- Nvidia Nemotron 3.5 Lightning Thinking: —

Math

AIME 2025
Competition mathematics, 2025 problems
- Nvidia Nemotron Super 49B: 7.7%
- Nvidia Nemotron 3.5 Lightning Thinking: —
AIME
Competition mathematics
- Nvidia Nemotron Super 49B: 19.3%
- Nvidia Nemotron 3.5 Lightning Thinking: —
MATH-500
Competition math problems
- Nvidia Nemotron Super 49B: 77.5%
- Nvidia Nemotron 3.5 Lightning Thinking: —

Legacy evaluations

Terminal-Bench Hard
Earlier terminal task suite
- Nvidia Nemotron Super 49B: 0.0%
- Nvidia Nemotron 3.5 Lightning Thinking: —
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- Nvidia Nemotron Super 49B: 12.0%
- Nvidia Nemotron 3.5 Lightning Thinking: 60.3%Top

Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart

curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nvidia/Llama-3.3-Nemotron-Super-49B-v1",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'

import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="nvidia/Llama-3.3-Nemotron-Super-49B-v1",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

nvidia

Other nvidia models, starting with the same family.