{
  "schemaVersion": "1.0",
  "updates": [],
  "companyNews": [],
  "id": "b703",
  "language": "en",
  "name": "Ragas",
  "originalName": "Ragas",
  "library": "tools",
  "category": "Model development",
  "categoryKey": "模型开发",
  "categoryKeys": [
    "模型开发"
  ],
  "categories": [
    "Model development"
  ],
  "entryPoints": [],
  "summary": "Evaluate RAG and LLM applications using traditional and model-based metrics, generating tests from source or production data.",
  "introduction": "Ragas provides metrics, test generation and feedback workflows for LLM applications. Developers organize inputs, answers and required context into examples, select traditional or model-based scoring and compare systems such as RAG. Source material or production data can generate scenario-specific tests. Evaluator models, prompts and sample distributions affect scores, making them comparative evidence rather than absolute certification.",
  "capabilities": [],
  "company": {
    "name": "VibrantLabs",
    "originalName": "VibrantLabs"
  },
  "officialUrl": "https://docs.ragas.io/en/stable/",
  "profileUpdatedOn": "2026-10-22",
  "sources": [
    {
      "kind": "product",
      "url": "https://docs.ragas.io/en/stable/"
    }
  ],
  "news": [],
  "links": {
    "interactive": "products/ragas--b703.html",
    "reading": "read/en/products/ragas--b703.html",
    "markdown": "read/en/products/ragas--b703.md",
    "json": "read/en/products/ragas--b703.json"
  }
}