Think Models

evroc Think Models is a Model-as-a-Service inference platform where evroc hosts leading open source AI models on NVIDIA H100 and B200 GPUs. You can either use our shared model endpoints or deploy a dedicated model instance yourself.

• Concepts — Understand the difference between Shared Models and Dedicated Model Instances, and when to choose each
• Supported Models — Browse the catalogue of available models with model cards, capabilities, and context limits
• OpenAI Compatibility — How to use the OpenAI-compatible API with cURL, the OpenAI Python client, and LangChain
• CLI — Manage model instances and API keys from the command line
• Inference API — Full OpenAPI specification for the evroc Think Models API