Skip to main content
NVIDIA NIM provides free API access to various open-source models optimized for NVIDIA infrastructure.

​
Overview

NVIDIA NIM (NVIDIA Inference Microservices) offers free access to a variety of open-source language models optimized for high-performance inference on NVIDIA GPUs.

​
Requirements

Phone Number Verification Required: You must verify your phone number to access NVIDIA NIM.

​
Rate Limits

Limit TypeFree Tier
Requests per minute40
Models tend to be context window limited on the free tier.

​
Available Models

NVIDIA NIM provides access to various open-source models including:
  • Llama models
  • Mistral models
  • Gemma models
  • Qwen models
  • And many more optimized open-source models

Browse All Models

Explore the full catalog of available models on NVIDIA NIM

​
API Usage

import openai

client = openai.OpenAI(
    base_url="https://integrate.api.nvidia.com/v1",
    api_key="YOUR_NVIDIA_API_KEY"
)

response = client.chat.completions.create(
    model="meta/llama-3.3-70b-instruct",
    messages=[
        {"role": "user", "content": "Hello, how are you?"}
    ],
    max_tokens=1024
)

print(response.choices[0].message.content)

​
Getting Started

1

Create Account

Visit build.nvidia.com
2

Verify Phone Number

Complete phone number verification
3

Generate API Key

Create an API key from your dashboard
4

Start Building

Use the OpenAI-compatible API to access models

​
Limitations

  • Context windows may be limited compared to paid tiers
  • 40 requests per minute across all models
  • Phone verification required for access

​
Additional Resources

NVIDIA NIM

Access the platform

Model Catalog

Browse available models

Build docs developers (and LLMs) love

Get started for freeTalk to us