modelparams.dev WebMCP Tools, Schemas & Agent Capabilities

modelparams.dev

live Unverified (seed import) AI & Agents 5 tools imperative implementation

A model's parameter count is its trained weights. API parameters are the request settings like temperature and top_p. Which one you want, and where to find it.

First seen 2026-11-05 · Last seen 2026-11-05 · Source: webmcp.com

Answer 5 · Act 0 · Transact 0 · Score: Not yet scored

Tool inspector

search_models

answer imperative

Search the LLM parameter catalog. Filter by free-text query, provider slug, required parameter path(s), API surface, and auth type (api_key | subscription | all). Also mirrors the query onto the page's visible filters. Returns matching model ids.

Page: /model-parameters-vs-api-parameters

Input schema

Property	Type	Required	Description
`query`	string	no	Free-text match on model name, provider, or id.
`provider`	string	no	Provider slug, e.g. anthropic, openai, deepseek.
`capability`	array	no	Parameter path(s) the model must support, e.g. thinking.type.
`auth`	string	no	Auth variant to include. Defaults to all.
`apiSurface`	string	no	API/SDK surface, e.g. openai-responses or anthropic-messages.
`limit`	number	no	Max models to return (default 25).

Raw JSON schema

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "description": "Free-text match on model name, provider, or id."
    },
    "provider": {
      "type": "string",
      "description": "Provider slug, e.g. anthropic, openai, deepseek."
    },
    "capability": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Parameter path(s) the model must support, e.g. thinking.type."
    },
    "auth": {
      "type": "string",
      "enum": [
        "all",
        "api_key",
        "subscription"
      ],
      "description": "Auth variant to include. Defaults to all."
    },
    "apiSurface": {
      "type": "string",
      "enum": [
        "openai-chat-completions",
        "openai-responses",
        "anthropic-messages",
        "google-generate-content",
        "amazon-bedrock-converse",
        "google-vertex-generate-content",
        "cohere-chat"
      ],
      "description": "API/SDK surface, e.g. openai-responses or anthropic-messages."
    },
    "limit": {
      "type": "number",
      "description": "Max models to return (default 25)."
    }
  }
}