OpenAI Guide

Use the OpenAI provider for OpenAI-hosted models. 3sparks Chat can fetch the models available to your API key and lets you choose a default model for each connection.

Add the connection

1. Open 3sparks Chat and go to Connections.
2. Tap + and choose OpenAI.
3. Leave the name blank for a generated name, or enter your own.
4. Confirm the base URL is api.openai.com, the route prefix is /v1/, and HTTPS is enabled.
5. Paste your OpenAI API key into API Token. Tokens are stored in Keychain.
6. Tap Create Connection.
7. Tap Fetch Models, choose a default model, and tap Done.

Reasoning, vision, and tokens

Models that support reasoning may expose extra controls in chat settings. Models that support image input can be used with photos, camera captures, pasted images, PDFs converted to images, and other image workflows.

The connection token limit controls how much context 3sparks Chat sends. Leave it unset for normal use, choose a preset when you know a model's context window, or set a custom limit when a provider rejects long prompts.

Troubleshooting

• Confirm the API key starts with sk- and has available credits.
• If a model disappears or becomes unavailable, tap Fetch Models again or use Switch Model in chat.
• If image messages fail, choose a model that supports vision.
• If responses stop early, check output token settings in chat settings.