S
Open-Source Only100+ ModelsOne Endpoint

ONE API

One endpoint. Every open-source model. Zero complexity.

Access 100+ open-source AI models through a single, unified API. No vendor lock-in, complete transparency.

View Documentation

Before ONE API

Multiple API keys to manage
Different request formats for each provider
Inconsistent response structures
Complex failover logic
Difficult cost tracking
Time-consuming integration updates
# Different code for each provider
if provider == "groq":
    client = Groq(api_key=groq_key)
elif provider == "together":
    client = Together(api_key=together_key)
# ... more complexity

With ONE API

Single API key for everything
Standard REST API format
Automatic failover and load balancing
Transparent usage tracking
No vendor lock-in
5-minute integration
# Same code for all models
client = OpenAI(
    api_key="your_one_api_key",
    base_url="https://theopensource.ai/v1"
)
# That's it!

Key Features

Everything you need to build with open-source AI

Standard REST API

Simple, well-documented REST API that works with any HTTP client. Easy integration with existing tools.

/v1/chat/completions

Multi-Provider Access

Access models from Groq, Together AI, and more. No need to manage multiple accounts or API keys.

Groq • Together AI • HuggingFace (soon)

Intelligent Routing

Automatic load balancing, cost optimization, and failover. Always get the best available provider.

99.9% uptime guaranteed

Usage Tracking

Detailed usage analytics and monitoring. Track token consumption and model performance in real-time.

Real-time dashboards • Usage reports

Enterprise-Grade Security

API key authentication, rate limiting, HTTPS encryption, and no data storage. Your data stays private.

SOC 2 compliant (coming soon)

Developer-Friendly

5-minute integration with extensive documentation, code examples, and SDKs for all major languages.

Python • Node.js • Go • Ruby • Java

Getting Started

Integrate in 5 minutes

1. Install SDK

pip install openai

2. Make Your First Request

from openai import OpenAI

client = OpenAI(
    api_key="your_one_api_key",
    base_url="https://theopensource.ai/v1"
)

response = client.chat.completions.create(
    model="llama-3.1-8b-instant",
    messages=[
        {"role": "user", "content": "Explain quantum computing"}
    ]
)

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

3. Advanced: Streaming

response = client.chat.completions.create(
    model="llama-3.1-8b-instant",
    messages=[{"role": "user", "content": "Write a story"}],
    stream=True
)

for chunk in response:
    print(chunk.choices[0].delta.content, end="")

100+ Open-Source Models

From small and fast to large and powerful

Small (7-8B)

Fast, cost-effective

llama-3.1-8b-instant
gemma-7b-it
mistral-7b-instruct

Medium (13-20B)

Balanced performance

mixtral-8x7b-instruct
deepseek-coder-33b
solar-10.7b-instruct

Large (70B+)

Maximum quality

llama-3.1-70b-versatile
qwen-2.5-72b-instruct
mixtral-8x22b

Use Cases

Built for every AI application

💬 Chatbots & Customer Support

Fast response times, cost-effective at scale, easy A/B testing

Recommended: llama-3.1-8b-instant

🔍 Content Generation

Large context windows, multiple models for different content types

Recommended: llama-3.1-70b

💻 Code Generation & Review

Specialized code models, fast iteration, cost-effective

Recommended: deepseek-coder-33b

📊 Data Analysis & Extraction

Structured output support, function calling, high accuracy

Recommended: qwen-2.5-72b

🌍 Multilingual Applications

Models trained on 50+ languages, translation, localization

Recommended: qwen-2.5-72b

🔐 Enterprise AI

No data retention, audit logging, usage controls, team management

Self-hosting available (coming Q2 2025)

Quick Integration

Get started in minutes

Simple Setup

Basic Usage
from openai import OpenAI

client = OpenAI(
    api_key="your_one_api_key",
    base_url="https://theopensource.ai/v1"
)

response = client.chat.completions.create(
    model="llama-3.1-8b-instant",
    messages=[{"role": "user", "content": "Hello!"}]
)

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

That's it! You're now connected to 100+ open-source models through a single endpoint.

Frequently Asked Questions

Ready to Get Started?

Join developers building with 100+ open-source AI models through ONE API

Explore Models

No credit card required • 5-minute setup • Open-source only