ChatWise × AI Code With
Use AI Code With in ChatWise: third-party relay setup guide
Use AI Code With inside the ChatWise desktop client. After you configure the built-in OpenAI provider, ChatWise lists the available models so you can pick one in chat. This page covers the setup steps, common errors, and FAQ.
Install ChatWise on macOS or Windows first, then finish provider setup in Settings. Enter the key only in ChatWise, never on this page.
- Provider
- OpenAI
- API Base URL
- https://api.aicodewith.ai/v1
- API Key
- sk-xxxxxxxx
Quick setup
Configure the built-in OpenAI provider in ChatWise settings. The client lists available models, and you can start chatting in about 3 minutes.
ChatWise is a macOS / Windows desktop AI client. Configure the built-in OpenAI provider once, and the fetched models are available in chat — no plugin and no config file needed.
Open Settings
Launch ChatWise, click the gear in the bottom left, then choose Settings. You can also press Cmd + ,.
Open ChatWise, click the gear in the bottom left, then choose Settings. You can also press Cmd + ,. Configure the built-in OpenAI provider
Open the Providers tab and select the built-in OpenAI entry on the left (you do not need to click + to add a custom provider). Paste your AI Code With API key and set API Base URL to https://api.aicodewith.ai/v1 (no trailing /).
Open Providers, pick the built-in OpenAI entry, paste your AI Code With API key, and set API Base URL to https://api.aicodewith.ai/v1. Pick a fetched model
Go back to the home screen and click the current model name in the top bar. ChatWise lists the models for this provider — pick one. You do not need to type Model IDs on this page.
After setup, ChatWise fetches compatible models for you. You do not need to add Model IDs on this page. View supported models
Back on the home screen, click the current model name in the top bar. ChatWise lists the models for this provider — pick one. You do not need to type Model IDs on this page. Send a test message
Pick a model and send a test message. A normal reply means the setup works — then open the AI Code With usage records and match the request by time and model name.
Pick a model and send a test message. A normal reply confirms the connection; match the call by time and model in AI Code With usage records.
How does ChatWise connect to AI Code With?
ChatWise ships with a built-in OpenAI provider. Connecting to AI Code With only needs two fields, then you pick a fetched model on the home screen:
- Open Settings
- On the home screen, click the gear in the bottom left, then choose Settings. You can also press Cmd + ,.
- Pick the built-in OpenAI provider
- Open the Providers tab at the top of Settings and select the built-in OpenAI entry on the left. You do not need to click + to add a custom provider.
- Enter connection details
- Paste your AI Code With API key and set API Base URL to https://api.aicodewith.ai/v1 (no trailing /).
- Pick a model and verify
- Go back to the home screen, click the current model name in the top bar, pick one from the list, and send a test message. A normal reply confirms the connection.
Flow: ChatWise client → built-in OpenAI provider → AI Code With API. Everything is configured inside ChatWise. No files to edit, and no Model IDs to type on this page.
Frequently asked questions
New to ChatWise?
ChatWise is a macOS / Windows desktop AI client. After you configure a third-party provider, you can use models in chat.
ChatWise official website
Open the official site and download the latest desktop build (use the Download button), then launch ChatWise.
Open the ChatWise websiteFull install guidance follows the ChatWise official docs; this site does not host ChatWise installers or images.
Common error troubleshooting
The most frequent problems fall into 4 categories.
401 / Invalid API Key
Symptom: Sending a message returns 401, Unauthorized, or Invalid API Key.
Cause: Usually the API key is invalid, mistyped, or copied with extra characters.
- Go back to the provider form and re-paste the API key (no extra spaces)
- Confirm the key is a valid AI Code With API key
- Save, then send a test message again
Empty model list / no model to pick
Symptom: Clicking the model name in the top bar shows an empty list, or only ChatWise defaults with no usable models.
Cause: The provider was not saved, the API Base URL is wrong, or ChatWise has not fetched /models yet.
- Confirm the built-in OpenAI provider is selected on the left, not another provider
- Confirm API Base URL is https://api.aicodewith.ai/v1 (with /v1, no trailing /)
- Close Settings and open the model list again
Base URL error
Symptom: Sending a message fails, or the model list does not load.
Cause: API Base URL has the wrong path, is missing /v1, or has a trailing /.
- Confirm the value is https://api.aicodewith.ai/v1
- Remove any trailing / slash
- Do not append /v1/models or other paths
No reply after sending a message
Symptom: You pick a model and send a test message, but there is no streamed reply or the UI keeps spinning.
Cause: The network is restricted, the provider is disabled, or the selected model is not available for the current key.
- Switch networks and retry
- Pick another model from the list and send another test message
- Check AI Code With usage records to see whether a call was made
Need the full guide?
Every config parameter, the complete model list, and more troubleshooting live in the AI Code With docs.
Open the AI Code With docs