Documentation

Supported Models

Using models from multiple providers with GATE/0

GATE/0 is designed to be model-agnostic, giving you the flexibility to access and switch between a wide range of large language models (LLMs) through a single, unified API. Whether you're using OpenAI’s latest models, Anthropic’s Claude, or models served via AWS Bedrock or Google Vertex AI, GATE/0 makes it seamless.

Built-in Support for Top Providers

GATE/0 natively supports multiple major model providers, including:

  • OpenAI
  • Azure OpenAI
  • Anthropic
  • Google Vertex AI
  • Amazon Bedrock
  • Together.ai
  • Cohere
  • Mistral
  • DeepSeek

This built-in integration means you don’t need to manage multiple SDKs or custom authentication logic. Just set your GATE/0 API key, specify the model, and you’re ready to go.

A Universal API for LLMs

With GATE/0’s universal API, your application code remains consistent, even as you switch between providers or models. This simplifies experimentation, reduces vendor lock-in, and allows for intelligent routing and fallback strategies.

For example, this code works regardless of the underlying model provider:

client = OpenAI(
    base_url="https://gateway.gate0.io/v1",
    api_key="your-gate0-api-key"
)

response = client.chat.completions.create(
    model="openai/gpt-4o",  # provider/model format
    messages=[{"role": "user", "content": "Hello!"}]
)

Specifying Model Providers

Some models are offered by multiple providers. For example:

  • gpt-4o is available via both openai and azure
  • claude-3-sonnet is available from anthropic, bedrock, and vertex

To avoid ambiguity, GATE/0 requires you to use the provider-prefixed model name, in the format:

{provider_slug}/{model_name}

The following table lists the available providers and their slugs:

ProviderSlug
OpenAIopenai
Azure OpenAIazure
Anthropicanthropic
Google Vertex AIvertex
Amazon Bedrockbedrock
Coherecohere
DeepSeekdeepseek

Examples:

  • openai/gpt-4o
  • azure/gpt-4o
  • bedrock/claude-3-sonnet
  • vertex/claude-3-haiku

This convention ensures clarity and precise routing of your requests to the correct backend provider.

Provider-prefixed model names

If you have only a single AI provider configure, you can use the model name without the provider prefix. However, it is recommended to use the provider prefix for consistency and clarity.

Why It Matters

  • Portability: Easily switch providers without changing your app logic.
  • Redundancy: Fall back to another provider if one is unavailable.
  • Cost Optimization: Choose the provider with the best pricing or performance for your needs.
  • Compliance: Use the provider that meets your region’s regulatory requirements.

Conclusion

GATE/0 gives you a unified, scalable way to interact with the growing ecosystem of LLMs — without the overhead of juggling multiple APIs. Just plug in the model name with the provider prefix and you’re ready to build.

Need help choosing the right model or provider for your use case? Contact us for personalized recommendations.