get_model_performance

Active

Tool of Tickerr - Live AI Tool Status & API Pricing

declared in 1.1.0

Get real-time API inference performance for LLM models - TTFT p50/p95 (time-to-first-token latency), throughput (tokens/sec), and 24-hour success rate. Use this alongside get_api_pricing to make routing decisions: cheapest model that is currently operational and fast enough for your latency budget. Covers anthropic, openai, google, groq, mistral, cerebras, cohere, xai, openrouter. Updated every 5 minutes via live API calls.

Parameters schema

{
  "type": "object",
  "properties": {
    "provider": {
      "type": "string",
      "description": "Provider name - \"anthropic\", \"openai\", \"google\", \"groq\", \"mistral\", \"cerebras\", \"cohere\", \"xai\", \"openrouter\"."
    }
  }
}

What this tool wraps· 4 endpoints

min confidence0.70 0.50

Data resourceOpenRouter API
OpenRouter · ai
name mention0.70
Data resourceCohere API
Cohere · ai
name mention0.65
Data resourceGroq API
Groq · ai
name mention0.65
Data resourceMistral AI API
Mistral AI · ai
name mention0.60

Parent server

Tickerr - Live AI Tool Status & API Pricing

https://github.com/imviky-ctrl/tickerr-mcp

2/7 registries

View full server →