API
Common questions related to our APIs and models
API Organization VerificationLearn how to verify your API organization
API Platform - Scale Tier for Existing Enterprise CustomersManage Scale Tier token capacity, purchases, authorized users, and usage for your enterprise API organization.
Agents API (beta) FAQGet started with managed agent sessions, execution environments, usage, and data controls.
Billing guide for the Reinforcement Fine Tuning APIHow billing works for the RFT API
Controlling the length of OpenAI model responsesLearn how to set output limits for OpenAI models using token settings, clear prompts, examples, and stop sequences.
Function calling in the Chat PlaygroundYou can use function calling in the OpenAI Playground.
Getting started with ZDR with Private Safety ProcessingSet up ZDR with PSP, validate your storage, and resolve onboarding issues.
Getting started with fine-tuningLearn how to prepare examples and find the guides for fine-tuning an OpenAI API model.
How can I change my API platform's org name?Changing your API platform's org name
How do I use stop sequences in the OpenAI API?Here is what you need to know about the stop sequence parameter in the OpenAI Chat Completions API
How to use the OpenAI API for Q&A or to build a chatbot?Using the Embeddings and Chat Completions API to create powerful question-answering applications
IP allowlisting for OpenAI APIRestrict which IP addresses can access your OpenAI API.
Migrating from GPT-4.x to GPT-6 LunaEvaluate GPT-6 Luna, compare cost and performance, and plan your migration from GPT-4.x.
Prompt management in PlaygroundHigh-quality prompts to kickstart every successful integration
Responses API support on Amazon BedrockLearn how AWS operates an OpenAI-compatible Responses API implementation for supported OpenAI models on Amazon Bedrock.
Sharing feedback, evaluation and fine-tuning data, and API inputs and outputs with OpenAIManage how your organization shares API feedback, evaluation data, fine-tuning data, and inputs or outputs with OpenAI.
Troubleshooting API Errors and LatencyThis article explains how to use the Service Health and Usage dashboards to troubleshoot common errors and latency issues when using the OpenAI API.
Using OpenAI models in CursorLearn how to use OpenAI models in Cursor with your own API key, the Codex IDE extension, or an AI gateway provider.
Using logit bias to alter token probability with the OpenAI APILearn how to use the logit bias parameter to modify model outputs
Common questions
Troubleshooting API rate limits and 429 errorsIdentify the cause of an API 429 error, reduce request bursts, and choose the right next step for rate, credit, or usage limits.
Troubleshooting API usage and spend limitsIdentify the limit or credit balance affecting your API requests and find the right next step.
Why am I getting different completions on Playground vs. the API?Troubleshooting discrepancies between completions
Why are Playground outputs misclassified as unsafe?Learn why Playground outputs may sometimes be incorrectly flagged as unsafe and how you can report misclassifications.
General FAQ
Admin and Audit Logs API for the API PlatformObtain audit logs and perform admin actions in your API Platform organization.
Best practices for prompt engineering with the OpenAI APIHow to give clear and effective instructions to OpenAI models
Do the OpenAI API models have knowledge of current events?Learn how model knowledge cutoffs and search tools affect access to current information.
Getting a Business Associate Agreement for the OpenAI APILearn how to accept a Business Associate Agreement and enable HIPAA compliance support for your API organization.
How do I start exploring the OpenAI API?Resources to get started building with the OpenAI API
Incorrect API key providedWhat to do if you get an Incorrect API key error on the OpenAI API
OpenAI Mutual TLS Beta ProgramConfigure mutual TLS for the OpenAI API, including client certificates, supported endpoints, and organization-level security settings.
Troubleshooting OpenAI account deactivationUnderstand why an OpenAI account may be deactivated and how to appeal a deactivation.
Where do I find my OpenAI API Key?Find, create, manage, and delete your OpenAI API keys while keeping them secure.
Why Did I Receive a Warning About My Account?Learn why OpenAI may issue an account warning and how to address policy or security concerns.
Security and API Key Safety
Best practices for API key safetyProtect your API keys with secure storage, access controls, monitoring, rotation, and IP restrictions.
Can I share my API key with my teammate/coworker?Learn why it’s best to keep API keys private and how to securely give your team access.
How can I view the users or organizations associated with an API key?Identify the user and organizations associated with an OpenAI API key by calling the account information endpoint.
How to Report Security Vulnerabilities to OpenAIOur policy and guidelines for testing and disclosure
Incorporating a safety identifierAdd a safety identifier to OpenAI API requests so abuse signals can be linked to users without sharing identifying information.
Text generation (Chat Completions and Completions)
GPT-3.5 and GPT-4, including function calling and vision
Batch API FAQBatch API endpoint for asynchronous batch processing
Function Calling in the OpenAI APIWhat is function calling and how does it work in the OpenAI API?
How can I use the Chat Completion API?Learn how to get started with the OpenAI Chat Completions API
How can I use the OpenAI API with text in different languages?Powerful text generation and understanding beyond English
Moving from Completions to Chat Completions in the OpenAI APILearn how to migrate from the legacy OpenAI Completions API to Chat Completions.
Understanding OpenAI API Billing and Usage
Fast mode FAQFrequently asked questions about Fast mode and the Ultrafast service tier.
Is the Moderation endpoint free to use?Learn about the cost of using the Moderation endpoint.
Is there an SLA for latency guarantees on the various engines?Learn how to contact OpenAI about latency requirements and check current or historical uptime on the OpenAI Status page.
Reviewing API usage and costsReview API activity in the Usage Dashboard, check token counts in API responses, and understand project filters, exports, and Scale Tier costs.
Understanding and counting tokensLearn how input, output, cached, and reasoning tokens affect API usage, model limits, and costs.