Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)
Intelligence
14.3
#359/523
Coding
7.6
#359/429
Math
7.7
#237/265
Speed
—
Pricing
—
Llama-3.3-Nemotron-Super-49B-v1 is a large language model (LLM) derived from Meta Llama-3.3-70B-Instruct. It’s post-trained for reasoning, chat, RAG, and tool calling, offering a balance between accuracy and efficiency (optimized for single H100). It underwent multi-phase post-training including SFT and RL (RLOO, RPO).
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) is NVIDIA’s advanced AI model designed for high-performance tasks. It features an Intelligence Index of 14.3, making it suitable for applications requiring substantial cognitive capabilities.
When to Use Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)
✓ Best For
- Natural language processing tasks.
- Data analysis and interpretation.
- Automated content generation.
✗ Not Ideal For
- Complex reasoning tasks.
- High-level mathematical computations.
How Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) Compares
Intelligence Index · Higher is better
MBZUAI Institute of Foundation ModelsNVIDIAAlibabaAnthropic
Benchmark Profile
Coding Index
AI21 LabsAnthropicNVIDIAAllen Institute for AI
Math Index
NVIDIAMetaAlibabaAmazon
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (13)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 14.3 | |
| Coding Index | 7.6 | |
| Math Index | 7.7 | |
| MMLU-Pro | 698% | |
| GPQA | 517% | |
| LiveCodeBench | 28% | |
| HLE | 35% | |
| SciCode | 22.9% | |
| IFBench | 39.5% | |
| LCR | 11.3% | |
| AIME | 19.3% | |
| AIME 2025 | 7.7% | |
| MATH 500 | 77.5% |
Data: Artificial Analysis · Updated: April 10, 2026
Frequently Asked Questions (15)
When was Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) released?
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) was released on March 18, 2025.
Who created Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)?
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) was created by NVIDIA.
How intelligent is Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)?
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) scores 14 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 13).
Is Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) a reasoning model?
No, Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
What input modalities does Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) support?
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) supports text input.
What output modalities does Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) support?
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) supports text output.
Can Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) process images?
No, Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) does not support image input. It can only process text.
Is Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) multimodal?
No, Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) is not multimodal. It only supports text input.
What is the context window of Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)?
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.
Is Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) open source?
Yes, Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
How many parameters does Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) have?
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) has 49 billion parameters.
What is the license for Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)?
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) is released under the NVIDIA Open Model License Agreement license. This license allows commercial use.
How does Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) perform on benchmarks?
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) achieves a score of 14 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
What is the knowledge cutoff date for Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)?
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) has a knowledge cutoff of December 2023. The model's training data includes information up to this date.
Is Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) available via API?
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) is an open weights model that can be self-hosted.
