Skip to content

Variable: ModelCapabilityOverrideSchema

const ModelCapabilityOverrideSchema: ZodObject<{ capabilities: ZodObject<{ aiGatewayCodex: ZodOptional<ZodBoolean>; anthropic: ZodOptional<ZodBoolean>; chat: ZodOptional<ZodBoolean>; customTools: ZodOptional<ZodBoolean>; embeddings: ZodOptional<ZodBoolean>; openResponses: ZodOptional<ZodBoolean>; parallelTools: ZodOptional<ZodBoolean>; reasoning: ZodOptional<ZodBoolean>; responses: ZodOptional<ZodBoolean>; streaming: ZodOptional<ZodBoolean>; structuredOutput: ZodOptional<ZodBoolean>; tools: ZodOptional<ZodBoolean>; webSearch: ZodOptional<ZodBoolean>; }, $strip>; model: ZodString; }, $strip>

Typed capability changes for one dynamically discovered model.

  • model: Model id or alias whose capabilities are overridden.
  • capabilities: Capability flags to merge onto the discovered model.
  • capabilities.responses (optional): True when the model can serve the Databricks Responses API.
  • capabilities.openResponses (optional): True when the model can serve OpenAI Responses.
  • capabilities.chat (optional): True when the model can serve Chat Completions.
  • capabilities.anthropic (optional): True when the model can serve Anthropic Messages.
  • capabilities.embeddings (optional): True when the model can serve embeddings.
  • capabilities.aiGatewayCodex (optional): True when the model is reachable through AI Gateway Codex.
  • capabilities.tools (optional): True when the model accepts tools.
  • capabilities.reasoning (optional): True when the model accepts reasoning effort.
  • capabilities.streaming (optional): True when the model can stream tokens.
  • capabilities.parallelTools (optional): True when the model can run multiple tools in one turn.
  • capabilities.customTools (optional): True when the model accepts custom tool definitions.
  • capabilities.structuredOutput (optional): True when the model can return structured output.
  • capabilities.webSearch (optional): True when the model can use web search.