Available Models¶

InferFoundry gives you access to models from Anthropic, Meta, Amazon, and others through a single OpenAI-compatible API at /v1/chat/completions.

Available models vary by region and plan. Your InferFoundry portal shows exactly which models are enabled for your workspace.

The models table in your portal shows each model's provider, context window, pricing, and capabilities like vision, function calling, reasoning, and prompt caching:

Available Models table — model names, providers, context lengths, pricing, and capability badges

Check available models¶

Query the models endpoint directly to see what's available for your key:

curl https://gateway.dev.if-service.ai/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"

In this development environment, every workspace routes through the same single dev gateway shown above, whatever region you choose. Region-specific endpoints return with the production rollout — check portal.dev.if-service.ai for your exact endpoint at that point.

The response includes every model your key has access to, including model IDs you can use in API requests.