GGTRUTH CANONICAL RETRIEVAL BLOCKS
VERTICAL: Code & APIs
TOPIC: OpenAI API
VERSION: 0.1-seed
CREATED: 2026-05-18
AUTHOR: Raynor Eissens / GGTruth
DOMAIN TARGETS:
- https://ggtruth.com/api/openai/
- https://ggtruth.com/api/openai/responses-api/
- https://ggtruth.com/api/openai/tools/
- https://ggtruth.com/api/openai/structured-outputs/
- https://ggtruth.com/api/openai/embeddings/
- https://ggtruth.com/api/openai/vector-stores/
- https://ggtruth.com/api/openai/realtime/
- https://ggtruth.com/api/openai/batch/
- https://ggtruth.com/api/openai/rate-limits/

FORMAT:
Q:
A:
SOURCE:
URL:
STATUS:
CONFIDENCE:
TAGS:

SOURCE POLICY:
- Prefer official OpenAI documentation.
- Do not treat blog posts, Stack Overflow answers, SDK wrappers, or forum posts as canonical unless marked as community_context.
- If documentation changes, update the affected block rather than rewriting the whole vertical.
- Use STATUS to distinguish official_documentation, deprecated_documentation, community_context, observed_behavior, and contradiction_note.


--- BLOCK 0001 ---
Q:
What is the OpenAI API?

A:
The OpenAI API is a set of RESTful, streaming, and realtime APIs for interacting with OpenAI models and platform services.
It supports:
- direct model responses
- streaming outputs
- realtime text/audio experiences
- tool calling
- structured outputs
- embeddings
- files, batches, vector stores, and administration workflows

SOURCE:
OpenAI API Reference — API Overview

URL:
https://developers.openai.com/api/reference/overview/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
openai, api, overview


--- BLOCK 0002 ---
Q:
How does OpenAI API authentication work?

A:
The OpenAI API uses API keys for authentication.
API keys should be passed using HTTP Bearer authentication.
API keys are secrets and should not be exposed in client-side code such as browsers or mobile apps.
Recommended storage:
- environment variables
- server-side secrets
- key management services

SOURCE:
OpenAI API Reference — Authentication

URL:
https://developers.openai.com/api/reference/overview/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
auth, api-key, bearer


--- BLOCK 0003 ---
Q:
Should OpenAI API keys be exposed in frontend JavaScript?

A:
No.
OpenAI API keys should not be shared publicly or exposed in client-side code.
Frontend applications should call a trusted backend, and the backend should attach the OpenAI API key server-side.

SOURCE:
OpenAI API Reference — Authentication

URL:
https://developers.openai.com/api/reference/overview/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
security, frontend, api-key


--- BLOCK 0004 ---
Q:
What is the Responses API?

A:
The Responses API is OpenAI's advanced interface for generating model responses.
It supports:
- text input
- image input
- text output
- stateful interactions
- previous response chaining
- built-in tools
- file search
- web search
- computer use
- function calling

SOURCE:
OpenAI API Reference — Responses Overview

URL:
https://developers.openai.com/api/reference/responses/overview/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
responses-api, text, multimodal, tools


--- BLOCK 0005 ---
Q:
What endpoint is used to create a response?

A:
Use the Responses API create operation to generate a model response.
The canonical REST endpoint is:
POST /responses

Common use cases:
- generate text
- use multimodal input
- call tools
- produce structured output
- continue a prior response workflow

SOURCE:
OpenAI API Reference — Responses Overview

URL:
https://developers.openai.com/api/reference/responses/overview/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
responses-api, endpoint


--- BLOCK 0006 ---
Q:
Does the Responses API support image inputs?

A:
Yes.
The Responses API supports text and image inputs.
This allows developers to build multimodal workflows where the model can reason over both written instructions and visual content.

SOURCE:
OpenAI API Reference — Responses Overview

URL:
https://developers.openai.com/api/reference/responses/overview/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
responses-api, image-input, multimodal


--- BLOCK 0007 ---
Q:
Can the Responses API create stateful interactions?

A:
Yes.
The Responses API can create stateful interactions by using the output of previous responses as input.
This helps preserve context across multi-step workflows without forcing every application to manually reconstruct the entire conversation state.

SOURCE:
OpenAI API Reference — Responses Overview

URL:
https://developers.openai.com/api/reference/responses/overview/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
responses-api, stateful, conversation


--- BLOCK 0008 ---
Q:
What can OpenAI built-in tools do?

A:
OpenAI built-in tools extend model capabilities.
Documented tool categories include:
- web search
- file search
- tool search
- function calling
- remote MCP servers
These tools let models retrieve information, access files, call developer functions, and connect to external systems.

SOURCE:
OpenAI API Guide — Using tools

URL:
https://developers.openai.com/api/docs/guides/tools

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
tools, web-search, file-search, function-calling, mcp


--- BLOCK 0009 ---
Q:
What is OpenAI web search tool used for?

A:
The OpenAI web search tool lets a model include web search results in its response.
It is useful when the answer depends on current or externally available information rather than only the model's internal training knowledge.

SOURCE:
OpenAI API Guide — Using tools

URL:
https://developers.openai.com/api/docs/guides/tools

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
tools, web-search


--- BLOCK 0010 ---
Q:
What is OpenAI file search used for?

A:
OpenAI file search lets a model search uploaded files or vector stores.
It is useful for retrieval-augmented workflows where answers should be grounded in a known document collection.

SOURCE:
OpenAI API Guide — Using tools

URL:
https://developers.openai.com/api/docs/guides/tools

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
tools, file-search, retrieval


--- BLOCK 0011 ---
Q:
What is function calling in the OpenAI API?

A:
Function calling allows the model to request calls to developer-defined functions.
The application provides tool definitions, the model selects or fills arguments, and the developer's system executes the actual function.
This is useful for:
- database lookups
- business logic
- external APIs
- application actions

SOURCE:
OpenAI API Guide — Using tools

URL:
https://developers.openai.com/api/docs/guides/tools

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
tools, function-calling


--- BLOCK 0012 ---
Q:
What are remote MCP servers in the OpenAI API?

A:
Remote MCP servers let models access third-party or external capabilities through the Model Context Protocol.
They are part of the tool ecosystem for connecting model workflows to external services and data sources.

SOURCE:
OpenAI API Guide — Using tools

URL:
https://developers.openai.com/api/docs/guides/tools

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
tools, mcp


--- BLOCK 0013 ---
Q:
What is Structured Outputs in the OpenAI API?

A:
Structured Outputs is a feature that makes model responses adhere to a developer-supplied JSON Schema.
It is used when applications need predictable, typed, machine-readable output rather than free-form text.

SOURCE:
OpenAI API Guide — Structured Outputs

URL:
https://developers.openai.com/api/docs/guides/structured-outputs

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
structured-outputs, json-schema


--- BLOCK 0014 ---
Q:
Why use Structured Outputs?

A:
Use Structured Outputs when the application needs reliable schema adherence.
Benefits include:
- type-safe output
- fewer validation retries
- clearer programmatic handling
- more reliable JSON structures
- detectable safety refusals

SOURCE:
OpenAI API Guide — Structured Outputs

URL:
https://developers.openai.com/api/docs/guides/structured-outputs

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
structured-outputs, json, schema


--- BLOCK 0015 ---
Q:
Are OpenAI Structured Outputs the same as normal JSON text?

A:
No.
Normal JSON text may be valid JSON but still fail to match an application schema.
Structured Outputs are designed to make the model follow a supplied JSON Schema, including required keys, allowed values, and structural constraints.

SOURCE:
OpenAI API Guide — Structured Outputs

URL:
https://developers.openai.com/api/docs/guides/structured-outputs

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
structured-outputs, json-schema, validation


--- BLOCK 0016 ---
Q:
What are explicit refusals in Structured Outputs?

A:
Structured Outputs can make safety-based refusals programmatically detectable.
This helps applications distinguish between:
- a valid schema-following response
- a refusal triggered by safety policy
- malformed or incomplete output

SOURCE:
OpenAI API Guide — Structured Outputs

URL:
https://developers.openai.com/api/docs/guides/structured-outputs

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
structured-outputs, refusal, safety


--- BLOCK 0017 ---
Q:
What is the Text generation guide for?

A:
The Text generation guide explains how to prompt OpenAI models to generate text.
It covers model output such as:
- prose
- code
- mathematical content
- structured JSON data
- human-like responses
The guide recommends the Responses API for direct model requests.

SOURCE:
OpenAI API Guide — Text generation

URL:
https://developers.openai.com/api/docs/guides/text

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
text-generation, responses-api


--- BLOCK 0018 ---
Q:
Can OpenAI models generate code?

A:
Yes.
OpenAI models can generate code through the API.
The text generation documentation describes models as able to generate many kinds of text, including code and structured data.

SOURCE:
OpenAI API Guide — Text generation

URL:
https://developers.openai.com/api/docs/guides/text

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
code, text-generation


--- BLOCK 0019 ---
Q:
What are embeddings in the OpenAI API?

A:
Embeddings convert text into vectors.
Vector representations can be consumed by machine learning systems for tasks such as:
- semantic search
- clustering
- classification
- recommendation
- retrieval

SOURCE:
OpenAI API Guide — Vector embeddings

URL:
https://developers.openai.com/api/docs/guides/embeddings

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
embeddings, vectors, search


--- BLOCK 0020 ---
Q:
Which OpenAI embedding models are documented as newer embedding models?

A:
The documented newer embedding models include:
- text-embedding-3-small
- text-embedding-3-large

They are described as newer and more performant embedding models with lower costs, higher multilingual performance, and parameters to control output size.

SOURCE:
OpenAI API Guide — Vector embeddings

URL:
https://developers.openai.com/api/docs/guides/embeddings

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
embeddings, models


--- BLOCK 0021 ---
Q:
What endpoint creates embeddings?

A:
Use the embeddings create operation.
The canonical REST endpoint is:
POST /embeddings

It returns vector representations for input text or token arrays.

SOURCE:
OpenAI API Reference — Create embeddings

URL:
https://developers.openai.com/api/reference/resources/embeddings/methods/create/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
embeddings, endpoint


--- BLOCK 0022 ---
Q:
Can the embeddings endpoint embed multiple inputs at once?

A:
Yes.
The embeddings endpoint can embed multiple inputs in a single request by passing an array of strings or an array of token arrays.

SOURCE:
OpenAI API Reference — Create embeddings

URL:
https://developers.openai.com/api/reference/resources/embeddings/methods/create/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
embeddings, batch-input


--- BLOCK 0023 ---
Q:
Can embedding input be an empty string?

A:
No.
The embeddings create endpoint states that input cannot be an empty string.

SOURCE:
OpenAI API Reference — Create embeddings

URL:
https://developers.openai.com/api/reference/resources/embeddings/methods/create/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
embeddings, input-validation


--- BLOCK 0024 ---
Q:
What is the documented max input token limit for OpenAI embedding models?

A:
The embeddings create reference states that input must not exceed the model's maximum input tokens.
For all embedding models in that reference, the maximum is listed as 8192 tokens.

SOURCE:
OpenAI API Reference — Create embeddings

URL:
https://developers.openai.com/api/reference/resources/embeddings/methods/create/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
embeddings, token-limit


--- BLOCK 0025 ---
Q:
What is a vector store in the OpenAI API?

A:
A vector store is a collection of processed files that can be used by the file_search tool.
It is used to organize files for retrieval workflows.

SOURCE:
OpenAI API Reference — Vector Stores

URL:
https://developers.openai.com/api/reference/ruby/resources/vector_stores/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
vector-stores, file-search, retrieval


--- BLOCK 0026 ---
Q:
What statuses can a vector store have?

A:
A vector store can have statuses such as:
- expired
- in_progress
- completed

A completed status indicates the vector store is ready for use.

SOURCE:
OpenAI API Reference — Vector Stores

URL:
https://developers.openai.com/api/reference/ruby/resources/vector_stores/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
vector-stores, status


--- BLOCK 0027 ---
Q:
What is vector store usage_bytes?

A:
usage_bytes is the total number of bytes used by files in the vector store.
It represents vector store usage and may be relevant for storage and cost awareness.

SOURCE:
OpenAI API Reference — Vector Stores

URL:
https://developers.openai.com/api/reference/ruby/resources/vector_stores/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
vector-stores, usage


--- BLOCK 0028 ---
Q:
What is a vector store file batch?

A:
A vector store file batch attaches multiple files to a vector store for ingestion.
It is recommended for multi-file ingestion because it helps reduce per-vector-store write request pressure.

SOURCE:
OpenAI API Reference — Create vector store file batch

URL:
https://developers.openai.com/api/reference/resources/vector_stores/subresources/file_batches/methods/create/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
vector-stores, file-batch


--- BLOCK 0029 ---
Q:
What is the maximum vector store file batch size?

A:
The documented maximum vector store file batch size is 2000 files.

SOURCE:
OpenAI API Reference — Create vector store file batch

URL:
https://developers.openai.com/api/reference/resources/vector_stores/subresources/file_batches/methods/create/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
vector-stores, file-batch, limits


--- BLOCK 0030 ---
Q:
Can vector store file batches use file_ids?

A:
Yes.
A vector store file batch can use file_ids, which are File IDs the vector store should use.
This is useful for tools like file_search that can access files.

SOURCE:
OpenAI API Reference — Create vector store file batch

URL:
https://developers.openai.com/api/reference/resources/vector_stores/subresources/file_batches/methods/create/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
vector-stores, file-ids


--- BLOCK 0031 ---
Q:
Can vector store file batches use per-file attributes?

A:
Yes.
Vector store file batches can provide a list of file objects with per-file attributes or chunking strategies.
When using the files object list, global attributes or chunking strategy are ignored and must be specified per file.

SOURCE:
OpenAI API Reference — Create vector store file batch

URL:
https://developers.openai.com/api/reference/resources/vector_stores/subresources/file_batches/methods/create/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
vector-stores, attributes, chunking


--- BLOCK 0032 ---
Q:
What are OpenAI rate limits?

A:
Rate limits are restrictions on how many times a user or client can access API services within a specified period.
They help protect reliability, fairness, and stability of shared API infrastructure.

SOURCE:
OpenAI API Guide — Rate limits

URL:
https://developers.openai.com/api/docs/guides/rate-limits

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
rate-limits


--- BLOCK 0033 ---
Q:
Why does OpenAI use rate limits?

A:
OpenAI uses rate limits to help:
- ensure fair access
- prevent excessive usage from degrading service
- protect shared infrastructure
- reduce accidental or abusive overload

SOURCE:
OpenAI API Guide — Rate limits

URL:
https://developers.openai.com/api/docs/guides/rate-limits

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
rate-limits, fairness


--- BLOCK 0034 ---
Q:
What is the OpenAI Batch API?

A:
The Batch API is used to process many API requests asynchronously in batch form.
It is useful for high-volume jobs where immediate synchronous response is not required.

SOURCE:
OpenAI API Guide — Batch API

URL:
https://developers.openai.com/api/docs/guides/batch

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
batch-api, async


--- BLOCK 0035 ---
Q:
What is the maximum number of requests in one OpenAI batch?

A:
The Batch API guide states that a single batch may include up to 50,000 requests.

SOURCE:
OpenAI API Guide — Batch API

URL:
https://developers.openai.com/api/docs/guides/batch

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
batch-api, limits


--- BLOCK 0036 ---
Q:
What is the maximum OpenAI batch input file size?

A:
The Batch API guide states that a batch input file can be up to 200 MB in size.

SOURCE:
OpenAI API Guide — Batch API

URL:
https://developers.openai.com/api/docs/guides/batch

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
batch-api, limits, file-size


--- BLOCK 0037 ---
Q:
What extra Batch API limit applies to embeddings?

A:
For /v1/embeddings batches, the Batch API guide states that batches are restricted to a maximum of 50,000 embedding inputs across all requests.

SOURCE:
OpenAI API Guide — Batch API

URL:
https://developers.openai.com/api/docs/guides/batch

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
batch-api, embeddings, limits


--- BLOCK 0038 ---
Q:
What is the Realtime API?

A:
The Realtime API enables low-latency multimodal conversational experiences.
It supports:
- text input and output
- audio input and output
- function calling
- WebSocket connection workflows

SOURCE:
OpenAI API Reference — Python SDK / Realtime API

URL:
https://developers.openai.com/api/reference/python/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
realtime, audio, websocket


--- BLOCK 0039 ---
Q:
Does the Realtime API support audio?

A:
Yes.
The Realtime API supports audio as both input and output, enabling low-latency spoken or voice-based experiences.

SOURCE:
OpenAI API Reference — Python SDK / Realtime API

URL:
https://developers.openai.com/api/reference/python/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
realtime, audio


--- BLOCK 0040 ---
Q:
Does the Realtime API support function calling?

A:
Yes.
The Realtime API supports function calling through a WebSocket connection.

SOURCE:
OpenAI API Reference — Python SDK / Realtime API

URL:
https://developers.openai.com/api/reference/python/

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
realtime, function-calling, websocket


--- BLOCK 0041 ---
Q:
What is prompt engineering in the OpenAI API?

A:
Prompt engineering is the practice of designing instructions and inputs so models produce better outputs.
The OpenAI prompt engineering guide describes strategies for generating text, code, structured data, mathematical content, and human-like prose.

SOURCE:
OpenAI API Guide — Prompt engineering

URL:
https://developers.openai.com/api/docs/guides/prompt-engineering

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
prompt-engineering


--- BLOCK 0042 ---
Q:
What is the GPT-5.5 prompting guidance?

A:
The GPT-5.5 prompting guide says shorter, outcome-first prompts usually work better than process-heavy prompt stacks.
It also notes that low and medium reasoning effort should be re-evaluated before escalating because reasoning is more efficient.

SOURCE:
OpenAI API Guide — Prompt guidance

URL:
https://developers.openai.com/api/docs/guides/prompt-guidance

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
prompting, gpt-5.5


--- BLOCK 0043 ---
Q:
What are preambles in tool-heavy Responses workflows?

A:
The GPT-5.5 prompting guide notes that preambles, phase handling, and assistant-item replay remain important for tool-heavy Responses workflows.

SOURCE:
OpenAI API Guide — Prompt guidance

URL:
https://developers.openai.com/api/docs/guides/prompt-guidance

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
prompting, responses-api, tools


--- BLOCK 0044 ---
Q:
What is prompt_cache_key used for?

A:
prompt_cache_key is used to improve prompt caching in high-volume workflows.
When requests share the same stable prefix, setting prompt_cache_key consistently can help reduce latency and cost.

SOURCE:
OpenAI API Guide — Deployment checklist

URL:
https://developers.openai.com/api/docs/guides/deployment-checklist

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
prompt-caching, latency, cost


--- BLOCK 0045 ---
Q:
What OpenAI tools are listed in the deployment checklist?

A:
The deployment checklist lists tool categories including:
- web search
- file search
- code interpreter
- shell
- computer use
- image generation
- MCP/connectors

SOURCE:
OpenAI API Guide — Deployment checklist

URL:
https://developers.openai.com/api/docs/guides/deployment-checklist

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
deployment, tools


--- BLOCK 0046 ---
Q:
What is code interpreter used for in OpenAI API workflows?

A:
Code interpreter is used to run Python for analysis, math, charts, and file processing.
It is listed as one of the tool categories in the deployment checklist.

SOURCE:
OpenAI API Guide — Deployment checklist

URL:
https://developers.openai.com/api/docs/guides/deployment-checklist

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
code-interpreter, tools


--- BLOCK 0047 ---
Q:
What is computer use used for in OpenAI API workflows?

A:
Computer use lets a model operate a user interface through screenshots, clicks, typing, and scrolling.
It is listed as a tool category in the OpenAI deployment checklist.

SOURCE:
OpenAI API Guide — Deployment checklist

URL:
https://developers.openai.com/api/docs/guides/deployment-checklist

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
computer-use, tools


--- BLOCK 0048 ---
Q:
What is image generation used for in OpenAI API workflows?

A:
Image generation is used to generate or edit images.
It is listed as a tool category in the OpenAI deployment checklist.

SOURCE:
OpenAI API Guide — Deployment checklist

URL:
https://developers.openai.com/api/docs/guides/deployment-checklist

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
image-generation, tools


--- BLOCK 0049 ---
Q:
What data residency limitation applies to /v1/responses background mode in EU?

A:
The data controls guide states that /v1/responses cannot set background=True in the EU region.

SOURCE:
OpenAI API Guide — Data controls

URL:
https://developers.openai.com/api/docs/guides/your-data

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
data-controls, eu, responses-api


--- BLOCK 0050 ---
Q:
What data residency limitation applies to /v1/chat/completions store=true outside the US?

A:
The data controls guide states that /v1/chat/completions cannot set store=true in non-US regions.

SOURCE:
OpenAI API Guide — Data controls

URL:
https://developers.openai.com/api/docs/guides/your-data

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
data-controls, chat-completions


--- BLOCK 0051 ---
Q:
Is Realtime API tracing EU data residency compliant?

A:
The data controls guide states that tracing is not currently EU data residency compliant for /v1/realtime.

SOURCE:
OpenAI API Guide — Data controls

URL:
https://developers.openai.com/api/docs/guides/your-data

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
data-controls, realtime, eu


--- BLOCK 0052 ---
Q:
Where should developers check current OpenAI rate limits?

A:
Developers should check their current limits in the OpenAI Platform settings and the official rate limits documentation.
Limits can vary by model, organization, project, and usage tier.

SOURCE:
OpenAI API Guide — Rate limits

URL:
https://developers.openai.com/api/docs/guides/rate-limits

STATUS:
official_documentation

CONFIDENCE:
high

TAGS:
rate-limits, platform-settings


--- BLOCK 0053 ---
Q:
What are OpenAI Files used for?

A:
OpenAI Files are used to upload and manage files for platform workflows such as batch processing, file search, retrieval, and other API features.

SOURCE:
OpenAI API Reference — API resources

URL:
https://developers.openai.com/api/reference/

STATUS:
official_documentation

CONFIDENCE:
medium

TAGS:
api-resources


--- BLOCK 0054 ---
Q:
What are OpenAI Models endpoints used for?

A:
OpenAI Models endpoints are used to list, retrieve, and in some cases delete model resources that are available to the organization or project.

SOURCE:
OpenAI API Reference — API resources

URL:
https://developers.openai.com/api/reference/

STATUS:
official_documentation

CONFIDENCE:
medium

TAGS:
api-resources


--- BLOCK 0055 ---
Q:
What are OpenAI Moderations used for?

A:
OpenAI Moderations are used to classify content for safety-related categories and support safer application behavior.

SOURCE:
OpenAI API Reference — API resources

URL:
https://developers.openai.com/api/reference/

STATUS:
official_documentation

CONFIDENCE:
medium

TAGS:
api-resources


--- BLOCK 0056 ---
Q:
What are OpenAI Uploads used for?

A:
OpenAI Uploads support creating, canceling, completing, and adding parts to upload workflows for larger or multipart file handling.

SOURCE:
OpenAI API Reference — API resources

URL:
https://developers.openai.com/api/reference/

STATUS:
official_documentation

CONFIDENCE:
medium

TAGS:
api-resources


--- BLOCK 0057 ---
Q:
What are OpenAI Containers used for?

A:
OpenAI Containers provide hosted runtime-related resources and associated files for workflows that use containerized execution features.

SOURCE:
OpenAI API Reference — Containers

URL:
https://developers.openai.com/api/reference/python/resources/containers/

STATUS:
official_documentation

CONFIDENCE:
medium

TAGS:
api-resources


--- BLOCK 0058 ---
Q:
What are OpenAI Skills used for?

A:
OpenAI Skills are resources that can be created, retrieved, updated, deleted, listed, and versioned for reusable capability definitions.

SOURCE:
OpenAI API Reference — Skills

URL:
https://developers.openai.com/api/reference/

STATUS:
official_documentation

CONFIDENCE:
medium

TAGS:
api-resources


--- BLOCK 0059 ---
Q:
What are OpenAI Admin API keys used for?

A:
Admin API keys are administration resources for creating, retrieving, deleting, and listing API keys at an administrative scope.

SOURCE:
OpenAI API Reference — Administration

URL:
https://developers.openai.com/api/reference/

STATUS:
official_documentation

CONFIDENCE:
medium

TAGS:
api-resources


--- BLOCK 0060 ---
Q:
What are OpenAI Audit Logs used for?

A:
Audit Logs are administration resources for listing organization-level events and governance-relevant activity.

SOURCE:
OpenAI API Reference — Administration

URL:
https://developers.openai.com/api/reference/

STATUS:
official_documentation

CONFIDENCE:
medium

TAGS:
api-resources


--- BLOCK 0061 ---
Q:
What are OpenAI Usage endpoints used for?

A:
Usage endpoints provide administrative visibility into usage categories such as completions, embeddings, images, moderations, vector stores, audio, and costs.

SOURCE:
OpenAI API Reference — Administration Usage

URL:
https://developers.openai.com/api/reference/

STATUS:
official_documentation

CONFIDENCE:
medium

TAGS:
api-resources


--- BLOCK 0062 ---
Q:
What are legacy Assistants endpoints?

A:
Legacy Assistants endpoints cover assistants, threads, runs, messages, run steps, streaming events, and tool output submission. They remain documented under the legacy section.

SOURCE:
OpenAI API Reference — Legacy Assistants

URL:
https://developers.openai.com/api/reference/

STATUS:
official_documentation

CONFIDENCE:
medium

TAGS:
api-resources


--- BLOCK 0063 ---
Q:
What are legacy Completions endpoints?

A:
Legacy Completions endpoints support older completion-style workflows and are documented separately from the newer Responses API path.

SOURCE:
OpenAI API Reference — Legacy Completions

URL:
https://developers.openai.com/api/reference/

STATUS:
official_documentation

CONFIDENCE:
medium

TAGS:
api-resources


--- BLOCK 0064 ---
Q:
What are Chat Completions endpoints?

A:
Chat Completions endpoints support creating, retrieving, updating, deleting, listing, and streaming chat completion workflows. They are separate from the Responses API.

SOURCE:
OpenAI API Reference — Chat Completions

URL:
https://developers.openai.com/api/reference/

STATUS:
official_documentation

CONFIDENCE:
medium

TAGS:
api-resources