AIbase
EN
Home
AI Models Finder
Llama 3.3 Nemotron Super 49B V1 (Reasoning)
L

Llama 3.3 Nemotron Super 49B V1 (Reasoning)

Llama-3.3-Nemotron-Super-49B-v1 is a large language model (LLM) derived from Meta Llama-3.3.70B-Instruct. It has undergone post-training for inference, chatting, RAG, and tool invocation, achieving a balance between accuracy and efficiency (optimized for a single H100). It has gone through multi-stage post-training, including SFT and RL (RLOO, RPO).
Intelligence(Medium)
Speed(Slow)
Input Supported Modalities
No
Is Reasoning Model
128,000
Context Window
131,072
Maximum Output Tokens
2023-12-31
Knowledge Cutoff
Go Compare

Pricing

- /M tokens
Input
- /M tokens
Output
- /M tokens
Blended Price

Quick Simple Comparison

Input

Output

Llama 3.3 Nemotron Super 49B v1 (Reasoning)
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)
Llama 3.3 Nemotron Super 49B v1

Basic Parameters

Llama 3.3 Nemotron Super 49B v1 (Reasoning)Technical Parameters
Parameter Count
49,900.0M
Context Length
128.00k tokens
Training Data Cutoff
2023-12-31
Open Source Category
Open Weights (Permissive License)
Multimodal Support
Text Only
Throughput
Release Date
2025-03-18
Response Speed
0 tokens/s

Benchmark Scores

Below is the performance of Llama 3.3 Nemotron Super 49B v1 (Reasoning) in various standard benchmark tests. These tests evaluate the model's capabilities in different tasks and domains.
Intelligence Index
51.16
Large Language Model Intelligence Level
Coding Index
27.98
Indicator of AI model performance on coding tasks
Math Index
77.1
Capability indicator in solving mathematical problems, mathematical reasoning, or performing math-related tasks
MMLU Pro
78.5
Massive Multitask Multimodal Understanding - Testing understanding of text, images, audio, and video
GPQA
64.3
Graduate Physics Questions Assessment - Testing advanced physics knowledge with diamond science-level questions
HLE
6.5
The model's comprehensive average score on the Hugging Face Open LLM Leaderboard
LiveCodeBench
27.7
Specific evaluation focused on assessing large language models' ability in real-world code writing and solving programming competition problems
SciCode
28.2
The model's capability in code generation for scientific computing or specific scientific domains
HumanEval
95.9
Score achieved by the AI model on the specific HumanEval benchmark test set
Math 500 Score
95.9
Score on the first 500 larger, more well-known mathematical benchmark tests
AIME Score
58.3
An indicator measuring an AI model's ability to solve high-difficulty mathematical competition problems (specifically AIME level)
AIbase
Empowering the Future, Your AI Solution Knowledge Base
English简体中文繁體中文にほんご
© 2025AIbase