JSON Examples Library

A library of 74 copy-ready JSON examples across 13 categories — from AI & LLM payloads and REST APIs to authentication, e-commerce, DevOps, and finance. Every example documents each field's type and meaning, shows real-world variants, and lists common use cases, so you learn the shape instead of just copying it. Open any example straight into the JSON Formatter, Validator, or type generators in one click — or generate realistic mock data from any example's structure. Use the search and category filters below to find exactly what you need.

AI & LLM

20
AI & LLM

OpenAI Chat Completion Response

OpenAI Chat Completions API response with choices, message, finish_reason, and usage.

openaillmchatgpt
AI & LLM

Anthropic Claude Messages Response

Claude Messages API response with content blocks, stop_reason, and token usage.

anthropicclaudellm
AI & LLM

LLM Tool / Function Call

Function-calling JSON: a tool definition plus the model's tool_call with arguments.

function callingtool usellm
AI & LLM

Embedding Vector Response

Embeddings API response with the vector array, model, and usage.

embeddingsvectorsemantic search
AI & LLM

RAG Document Chunk

A RAG chunk with text, source metadata, embedding reference, and relevance score.

ragretrievalchunk
AI & LLM

Chat Conversation History

Ordered messages array with system, user, and assistant roles for LLM chats.

chatconversationmessages
AI & LLM

AI Agent Run Trace

Agent run with goal, reasoning steps, tool calls, observations, and final output.

ai agentagentstrace
AI & LLM

MCP Message (Model Context Protocol)

Model Context Protocol (MCP) JSON-RPC 2.0 message: tools/list and tools/call.

mcpmodel context protocoljson-rpc
AI & LLM

LLM Structured Output (JSON Schema)

Structured output request with a JSON Schema and the model's schema-valid response.

structured outputjson modeopenai
AI & LLM

Fine-Tuning Dataset Example (JSONL)

A single fine-tuning training example in the chat messages JSONL format.

fine-tuningjsonltraining data
AI & LLM

Vector Database Search Query

Vector DB query with topK and filter, plus a ranked matches response.

vector databasepineconeqdrant
AI & LLM

Speech-to-Text Transcription (Whisper)

Speech-to-text response with text, language, duration, and timestamped segments.

whisperspeech to texttranscription
AI & LLM

AI Image Generation Request & Response

Text-to-image generation request and response with prompt, size, and image URLs.

image generationdall-etext to image
AI & LLM

Prompt Template (Reusable)

Reusable prompt template with variables, model params, and few-shot examples.

promptprompt templateprompt engineering
AI & LLM

LangChain / LlamaIndex Document

A LangChain/LlamaIndex Document with page_content and metadata for RAG pipelines.

langchainllamaindexdocument
AI & LLM

AI Agent Tool Registry

A registry of agent tools, each with a name, description, and JSON Schema parameters.

ai agenttoolsfunction calling
AI & LLM

LLM Evaluation / Benchmark Result

LLM eval run with per-case scores, metrics, and aggregate pass rate.

llm evalevaluationbenchmark
AI & LLM

Content Moderation Response

Moderation API response with flagged status, category flags, and scores.

moderationcontent moderationtrust and safety
AI & LLM

RAG Answer with Citations

A grounded RAG answer with an array of cited sources and confidence.

ragcitationsgrounding
AI & LLM

OpenAI Chat Completion Response

OpenAI chat completion response JSON with message content, finish_reason, and token usage.

openaillmchat completion

API Design

10

Web & Standards

4

Authentication

2

Data

14
Data

Address

Postal address JSON with structured fields for street, city, state, country, and coordinates.

addresspostalshipping
Data

Geolocation

GPS coordinates with accuracy, altitude, speed, and reverse-geocoded address.

geolocationGPScoordinates
Data

App Config File

Backend config with database connection, auth settings, feature flags, and logging.

configsettingsbackend
Data

Weather Data

Current weather and forecast with temperature, wind, humidity, and UV index.

weatherforecastAPI
Data

HTTP Status Codes

Reference dataset of 12 HTTP status codes with category, text, and description.

HTTPstatus codesREST API
Data

Programming Languages

Reference JSON for 8 popular languages with year, creator, paradigm, and use case.

programming languagesreferencedataset
Data

JSON Form Schema

Dynamic form definition JSON with field types, validation, selects, and layout.

form schemadynamic formsJSON Schema
Data

Country Codes (ISO 3166-1)

ISO 3166-1 country codes with alpha-2, alpha-3, numeric, continent, and calling code.

ISO 3166country codesalpha-2
Data

Currency Codes (ISO 4217)

ISO 4217 currency codes with symbols, minor units, and sample countries.

ISO 4217currency codesfinance
Data

MIME Types Reference

Common MIME types with file extensions, categories, and usage descriptions.

MIME typeContent-TypeHTTP headers
Data

IANA Timezones Reference

Major IANA timezone IDs with UTC offsets, DST offsets, and regional metadata.

IANA timezonetime zoneUTC offset
Data

US States (All 50 + DC)

All 50 US states and DC with postal codes, capitals, regions, and statehood years.

US statesUnited Statesaddress validation
Data

India States & Union Territories

All 28 Indian states and 8 Union Territories with codes, capitals, regions, and ISO codes.

India statesIndian statesGSTIN
Data

MongoDB Document

MongoDB document JSON with ObjectId, embedded address sub-document, and referenced order ids.

mongodbdocumentnosql

DevOps

6

E-commerce

3

Business

5

Finance

2

Content

3

Social

2

IoT & Sensors

2

Healthcare

1

How to use the JSON Examples Library

Every developer hits the same wall: you know roughly what a piece of JSON should look like — a user object, an API error, an LLM tool call — but you want a correct, complete reference instead of guessing field names and nesting. That's what this library is for. Each example is a hand-written, real-world structure you can copy in one click and adapt to your own data.

Examples go beyond the raw snippet. Every page documents each field's type, whether it's required, and what it means, so you understand the shape rather than blindly copying it. Most include several variants — a request vs. a response, a single item vs. a collection, a minimal vs. a full version — plus real use cases and FAQs.

The fastest workflow is to open an example and click Format, Validate, Explore tree, or Generate TypeScript — the exact JSON loads straight into that tool. You can also paste your own data into the JSON Formatter, Validator, or JSON to TypeScript generator at any time.

Building AI features? Start with the AI & LLM category at the top — it covers the exact JSON shapes you parse when working with OpenAI, Anthropic Claude, embeddings, vector search, RAG, agents, and the Model Context Protocol (MCP). For a guided tour, read JSON for AI & LLM Engineering.

Frequently asked questions

A JSON example is a realistic, copy-ready snippet showing how a particular kind of data — an API response, a config file, an LLM payload — is structured in JSON. Use it as a starting point: copy it, swap in your values, and validate it. Every example here also documents each field's type and meaning, shows real-world variants, and lets you open the JSON directly in a formatter, validator, or type generator.

Yes. Every example is free to copy and use in your own projects, docs, tests, or mock servers with no attribution required. They're hand-written reference structures, not copyrighted data.

Yes — the AI & LLM category is featured at the top and covers OpenAI chat completions, Anthropic Claude messages, tool/function calling, MCP, embeddings, vector search, RAG chunks and citations, agent traces, fine-tuning datasets, structured output, evals, moderation, Whisper, and image generation.

Open any example and click Format, Validate, Explore tree, or Generate TypeScript — the exact JSON loads straight into that tool, no copy-paste needed. You can also paste your own JSON into the JSON Formatter or JSON Validator at any time.

Many real structures have common variations — a request body vs. a response, a single object vs. a collection, a minimal vs. a full version. Each example page lists these as labelled variants so you can see how the shape changes across scenarios.

Got a JSON example to validate?

Paste any JSON into JSONKit's tools — format, validate, minify, or convert it in seconds.