Nemotron-4 340B

Open Source
By NVIDIAReleased 2025-06-01128K context340B parameters
Open SourceAPI Available

Benchmark Scores

82.2/100
MMLU85.5
HumanEval82
Math78.5
Reasoning84
Coding81

About Nemotron-4 340B

Nemotron-4 340B is NVIDIA's large open-source model optimized for synthetic data generation and running on NVIDIA hardware. It excels at creating high-quality training data and instruction following.

Strengths

  • +Optimized for NVIDIA hardware
  • +Strong at synthetic data generation
  • +Good instruction following
  • +Open source

Weaknesses

  • -Requires NVIDIA infrastructure
  • -Less versatile than general models
  • -Large compute requirements

Pricing

Per 1M tokens

Input$0.00
Output$0.00

Quick Facts

Context Window
128K
Parameters
340B
Release Date
2025-06-01
Category
Open Source

Compare Models

See how Nemotron-4 340B stacks up against other models

More from NVIDIA

Try Nemotron-4 340B on Vincony

Compare Nemotron-4 340B with 400+ other AI models side-by-side. Run AI debates, generate images and videos, and find the best model for your task.

Visit Vincony.com