Evals · Schema

EvalDataset

A collection of EvalCases plus metadata describing its provenance, license, splits, and the task it targets. Maps directly to Hugging Face datasets (MMLU, HumanEval, GAIA), LangSmith datasets, Braintrust datasets, and DeepEval EvaluationDatasets.

EvalsLLM EvaluationAI QualityBenchmarksLLM as a JudgeObservabilityAgent EvaluationRAG EvaluationTest-Driven AI

Properties

Name Type Description
id string
name string
description string
version string
task string What the dataset is for.
source string Canonical URL where the dataset is published.
license string
splits object Named subsets (train/dev/test, or domain-specific splits).
case_count integer Total number of cases across all splits.
tags array
created string
modified string
View JSON Schema on GitHub

JSON Schema

evals-dataset-schema.json Raw ↑
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://raw.githubusercontent.com/api-evangelist/evals/refs/heads/main/json-schema/evals-dataset-schema.json",
  "title": "EvalDataset",
  "description": "A collection of EvalCases plus metadata describing its provenance, license, splits, and the task it targets. Maps directly to Hugging Face datasets (MMLU, HumanEval, GAIA), LangSmith datasets, Braintrust datasets, and DeepEval EvaluationDatasets.",
  "type": "object",
  "properties": {
    "id": {
      "type": "string",
      "example": "ds_support_faq_2026q2"
    },
    "name": {
      "type": "string",
      "example": "Customer Support FAQ 2026 Q2"
    },
    "description": {
      "type": "string",
      "example": "Curated and human-verified set of 240 customer-support questions and reference answers for evaluating the support RAG pipeline."
    },
    "version": {
      "type": "string",
      "example": "2026.05.0"
    },
    "task": {
      "type": "string",
      "description": "What the dataset is for.",
      "enum": ["qa", "rag", "code_generation", "summarization", "classification", "agent_task", "safety", "multi_turn_dialogue", "knowledge", "reasoning"],
      "example": "rag"
    },
    "source": {
      "type": "string",
      "format": "uri",
      "description": "Canonical URL where the dataset is published.",
      "example": "https://huggingface.co/datasets/cais/mmlu"
    },
    "license": {
      "type": "string",
      "example": "MIT"
    },
    "splits": {
      "type": "object",
      "description": "Named subsets (train/dev/test, or domain-specific splits).",
      "additionalProperties": {
        "type": "object",
        "properties": {
          "count": { "type": "integer", "example": 240 },
          "uri": { "type": "string", "format": "uri" }
        }
      },
      "example": {
        "train": { "count": 180 },
        "test": { "count": 60 }
      }
    },
    "case_count": {
      "type": "integer",
      "description": "Total number of cases across all splits.",
      "example": 240
    },
    "tags": {
      "type": "array",
      "items": { "type": "string" },
      "example": ["rag", "support", "en-US"]
    },
    "created": {
      "type": "string",
      "format": "date-time"
    },
    "modified": {
      "type": "string",
      "format": "date-time"
    }
  },
  "required": ["id", "name", "task"]
}

Work with this as data

Every JSON Schema here is available over the APIs.io API and to AI agents over MCP.

MCP server

One button, every client — Claude, Cursor, VS Code and the rest.

https://apis.io/mcp

Tools for schemas

4 MCP tools reach this
  • find_json_schemasBrowse and filter every JSON Schema in the catalog.
  • apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.
  • resolveTurn a domain, URL or GitHub org into the provider it belongs to.
  • find_cohortsEvery scored population of providers in the catalog.
All 92 tools →

Call it yourself

curl for this page
This JSON Schema
curl "https://apis.io/api/v1/json-schemas/evals-dataset"
All schemas
curl "https://apis.io/api/v1/json-schemas?limit=25"

Discovery needs no key. Ratings and market analysis are Pro.

Get an API key

Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. See our Privacy Policy and Terms.

A second provider on the same verified email joins the account you already have.