artificial-analysis-mcp

Artificial Analysis MCP Server

Artificial Analysis pricing, speed measurements, and benchmark results become available to AI assistants through MCP. It supports model selection with creator-filtered listings, sortable metrics, and detailed lookups for individual models.

Server facts

artificial-analysis-mcp

Project

License
Not specified

Tech

Runtime / language
Not specified
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
9
Last commit
Data as of

Tags

  • analysis
  • benchmark
  • cost
  • llms
  • performance
  • speed

Overview

Model selection involves comparing cost, response speed, and benchmark performance. artificial-analysis-mcp brings those measurements from Artificial Analysis into assistant workflows. Its list_models tool retrieves the model catalog, with optional creator filtering, result limits, and sorting. Users can narrow the list to a provider such as Anthropic or Google, then order results by pricing, generation speed, initial response latency, benchmark performance, or release date. Ascending and descending ordering support different comparisons, including finding lower-priced models or examining stronger benchmark results.

For a closer look, get_model accepts a model name or slug and returns its detailed record. That includes separate input, output, and blended pricing, alongside output-token throughput and time to first token. Benchmark coverage includes measures for intelligence, coding, and math, plus GPQA, MMLU-Pro, and LiveCodeBench. This lets an assistant move from a broad comparison to a specific model's cost and performance details. Configuration requires an Artificial Analysis API key; the documented setup includes Claude Code, and the listed transport is stdio.

artificial-analysis-mcp setup notes

  • artificial-analysis-mcp records stdio only; confirm a network transport upstream before planning remote access.
  • artificial-analysis-mcp has no recorded license; confirm reuse terms in its repository.

Related MCP servers