AI & Machine Learning

modelparams.dev

Provides tools to validate and inspect LLM model parameters, list models, and retrieve provider-specific model information.

1 endpoint4 known toolsFirst detected August 30, 2026Last detected August 30, 2026

ENDPOINT 1

https://modelparams.dev/mcp

No auth detected

MCP server metadata

Name: modelparams
Version: 1ae09d3

Capabilities

tools.listChanged

Server instructions

Answers what parameters an LLM accepts, using the modelparams.dev catalog of 382 models. Call validate_model_params before issuing any LLM request that sets non-default parameters, and when diagnosing a 400 from a provider — it catches unsupported parameters and invalid combinations that a provider would reject or silently ignore.

Known tools 4

validate_model_params

Check a set of LLM request parameters against what a model actually accepts, before calling the provider.

Inferred read-only

get_model_params

List every parameter a model accepts, with its type, allowed range or enum values, default, and any conditional rules governing when it applies.

Inferred read-only

list_models

List model ids in the modelparams.dev catalog, optionally filtered by provider or a substring.

Inferred read-only

list_provider_models

Everything one provider serves: its models, the wire id each one needs, their lifecycle status, and the parameter surfaces they expose — types, ranges, enum values, defaults, and conditional rules — in a single call.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.modelparams]
url = "https://modelparams.dev/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "modelparams": {
      "type": "http",
      "url": "https://modelparams.dev/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: modelparams
Remote MCP URL: https://modelparams.dev/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "modelparams": {
      "url": "https://modelparams.dev/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "modelparams": {
      "type": "http",
      "url": "https://modelparams.dev/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "modelparams",
  "transport": "streamable-http",
  "url": "https://modelparams.dev/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.