Whatever message this page gives is out now! Go check it out!
OpenAI LLM configuration- field reference
Last update:
May 18, 2026
This document describes settings for the OpenAI provider in the LLM configuration UI (basic connection, model parameters, proxy, and advanced options).
Behavior notes
- Leaving numeric or optional text fields blank uses the provider or integration default for that parameter.
- API key and logging copy includes inline security guidance as shown in the UI.
Basic configuration
Field | Description |
|---|---|
Config alias | Required. Unique name for this configuration. Use only letters, numbers, hyphens, and underscores (for example gpt4-prod). |
Provider | OpenAI. The provider for this saved configuration. |
Model name | Required. The model identifier OpenAI expects for this request path (for example gpt-4o). Use the exact id from OpenAI model documentation. |
API key | Optional. Leave blank to supply the key at runtime. If you enter a key, it is stored with this configuration—treat it as a secret and follow your organization’s access and rotation rules. |
Base URL | Optional. Use the default OpenAI endpoint unless your integration uses a custom gateway, proxy URL, or compatible API base. When set, use the root URL your deployment expects. |
Organization ID | Optional. OpenAI organization id when your account uses organization scoping. Leave blank if you do not use organizations. |
Project ID | Optional. OpenAI project id when your workflow uses project-scoped access. Leave blank if not used. |
---
Model parameters (sampling and output)
Field | Description |
|---|---|
Temperature | Leave blank to use the default. When set, use 0 or greater. Higher values increase randomness; typical range is 0–2 per OpenAI APIs. |
Max tokens | Leave blank to use the default. When set, enter a positive integer for the maximum number of tokens in the model response. |
Top P | Leave blank to use the default. When set, use 0 or greater (often 0–1). Nucleus sampling; ignored for some models if unsupported. |
Presence penalty | Leave blank to use the default. When set, use 0 or greater. Reduces likelihood of repeating topics already present in the context. |
Frequency penalty | Leave blank to use the default. When set, use a value in the range OpenAI documents for your API version (commonly -2.0 to 2.0). Reduces repetition of the same tokens. |
Seed | Leave blank to use the default. When set, use a positive integer for more deterministic sampling when the model supports it; behavior may still vary. |
Response format | Default or structured output when your integration requests JSON or other formats OpenAI supports for the selected model. Leave default if you do not need structured responses. |
Stop sequences | Optional. Comma-separated sequences that stop generation (for example END or an escaped newline your API accepts). |
---
OpenAI-only parameters
These appear for OpenAI (and sometimes Azure OpenAI) in the full parameter matrix; include them in help when your UI exposes them.
Field | Description |
|---|---|
Logit bias (JSON) | Optional. JSON map of token ids to bias values. Use only when you rely on logit bias; invalid JSON or ids cause request errors. |
Metadata (JSON) | Optional. OpenAI-specific metadata payload. Must be valid JSON per your API usage. |
Parallel tool calls | When enabled, allows parallel tool calls if your integration uses tools with the OpenAI API. Disable if your client expects sequential tool handling. |
Strict JSON schema | When enabled, requests stricter structured output behavior where the API supports it. Use with compatible models and schemas only. |
---
Proxy settings
Field | Description |
|---|---|
Proxy host | Optional. HTTP proxy hostname or IP. If you set a host, you must set a proxy port. |
Proxy port | Optional. Integer from 1 through 65535. Required when proxy host is set. |
Proxy username | Optional. Proxy authentication user name. Leave blank if the proxy does not require credentials. |
Proxy password | Optional. Proxy password. Stored credentials are sensitive—follow your organization’s secret-handling rules. Leave blank if not required. |
---
Advanced settings
Field | Description |
|---|---|
Max retries | Leave blank to use the default. When set, enter a positive integer for retry attempts after failed requests. |
Timeout (seconds) | Leave blank to use the default. When set, enter a positive integer for the maximum wait time per request. |
User ID | Optional. End-user identifier OpenAI accepts for abuse monitoring and safety (maximum 256 characters). Use a stable id per end user where policy allows; leave blank if not required. |
Log API requests | When enabled, request details are written to logs. Requests can include prompts, headers, and identifiers—use only while troubleshooting and disable afterward. Follow your organization’s logging and privacy requirements. |
Log API responses | When enabled, response details are written to logs. Responses can contain generated text and personal data—use only while troubleshooting and disable afterward. Follow your organization’s logging and privacy requirements. |
---
Not used for OpenAI (reference)
The shared six-provider form may hide or omit fields that apply only to other providers. For OpenAI, you typically do not show: Top K, Repeat penalty (Ollama-style), Anthropic beta / caching / thinking, Gemini code execution, or Azure deployment name / API version fields—unless your product maps multiple providers to one screen.
---
Related reference
- Parameter inventory and validation rules: see
LLM_PROVIDER_PARAMETERS.mdin your documentation or repo (OpenAI column: Y = supported, -- = not shown for that provider).
We're glad. Tell us how this page helped.
We're sorry. Can you tell us what didn't work for you?
Share additional feedback to help us improve.
0/255
|
