# LLMCapabilities Capabilities and features supported by an LLM service ```python from goodmem.models import LLMCapabilities ``` - `supports_chat` (`bool | None`, optional): Supports conversational/chat completion format with message roles JSON: `supportsChat`. - `supports_completion` (`bool | None`, optional): Supports raw text completion with prompt continuation JSON: `supportsCompletion`. - `supports_function_calling` (`bool | None`, optional): Supports function/tool calling with structured responses JSON: `supportsFunctionCalling`. - `supports_system_messages` (`bool | None`, optional): Supports system prompts to define model behavior and context JSON: `supportsSystemMessages`. - `supports_streaming` (`bool | None`, optional): Supports real-time token streaming during generation JSON: `supportsStreaming`. - `supports_sampling_parameters` (`bool | None`, optional): Supports sampling parameters like temperature, top_p, and top_k for generation control JSON: `supportsSamplingParameters`. [Python](../../python.md)