DeepChat Γ AI Code With
Use AI Code With in DeepChat: Third-Party API / Relay Setup
DeepChat is an open-source desktop AI client for managing providers, models, and chats locally, including custom OpenAI-compatible endpoints.
Download and install DeepChat first, then add a custom provider in Settings. Enter the key only in the app, never on this page.
- API Type
- OpenAI Completions
- API URL
- https://api.aicodewith.ai/v1
- API Key
- sk-xxxxxxxx
- Model ID
- claude-sonnet-4-6
Quick setup
In DeepChat Settings, add a custom provider: choose OpenAI Completions, fill the API URL and key, then enable models. About 3 minutes.
DeepChat is a desktop client. Open Settings Overview, go to Providers, add a custom provider, fill API type, URL, and key, then connect and enable models.
Open Providers
Open DeepChat Settings, then click the Providers card on Settings Overview.
Open DeepChat Settings, then click the Providers card on Settings Overview to open provider setup. Add a custom provider
Click Add custom provider. Set the name to aicodewith and the API type to OpenAI Completions. Set the API URL to https://api.aicodewith.ai/v1 (include /v1, do not append /chat/completions), then paste the API key.
Click Add custom provider. Set the name to aicodewith, API type to OpenAI Completions, API URL to https://api.aicodewith.ai/v1, paste the API key, then click Connect and load models. Connect and enable models
Click Connect and load models. Search the list and turn on the models you need; if one is missing, click Add model and paste a Model ID from this page.
Copy a Model ID
DeepChat fetches the model list after you connect. Copy a Model ID below, then search the list and enable the toggle; if it is missing, click Add model and paste it.
Pick a family, then copy a Model ID. The API URL and type stay the same for every model.
Selected
- Provider
- Claude
- Model ID
- claude-sonnet-4-6
The full catalog follows the AI Code With docs and pricing page. Example IDs in screenshots only show where to enable a model; use an ID from this page.
View the full model listAfter a successful connection DeepChat loads the model list. Enable the models you need; if one is missing, click Add model and paste a Model ID from this page. Send a test message
Return to chat, pick an enabled model at the bottom, and send a test message. A normal reply confirms the connection β then match the call by time and model in AI Code With usage records.
Return to chat, pick an enabled model at the bottom, 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 DeepChat connect to AI Code With?
DeepChat talks to an OpenAI-compatible API through a custom provider. Connecting to AI Code With means adding a provider, filling the URL and key, then enabling models:
- Open Providers
- Open DeepChat Settings, then click the Providers card on Settings Overview.
- Add a custom provider
- Click Add custom provider. Set the name to aicodewith and the API type to OpenAI Completions (not a native Anthropic type).
- Fill the URL and key
- Set the API URL to https://api.aicodewith.ai/v1 (include /v1, do not append /chat/completions). Paste your AI Code With API key.
- Connect and enable models
- Click Connect and load models. Turn on the models you need; if one is missing, click Add model and paste a Model ID from this page.
- Send a test message
- Return to chat, pick an enabled model at the bottom, and send a test message. A normal reply confirms the connection.
Flow: DeepChat app β custom provider (OpenAI Completions + API URL + key) β AI Code With API. Everything is configured in Settings. No files to edit.
Frequently asked questions
New to DeepChat?
DeepChat is an open-source, local-first desktop AI client for macOS, Windows, and Linux.
DeepChat website
Download the installer for your OS from the official site. After install, open DeepChat and go to provider setup from Settings Overview.
Open the DeepChat websiteFull install guidance follows the DeepChat official site; this site does not host the installer.
Common error troubleshooting
The most frequent problems fall into 4 categories.
401 / Invalid API key
Symptom: Connect and load models, or a chat 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 custom provider form and re-paste the API key (no extra spaces)
- Confirm the key is a valid AI Code With API key
- Click Connect and load models again
404 / Wrong API URL
Symptom: The connection fails, or the request hits .../v1/chat/completions/chat/completions.
Cause: The API URL was set to the full chat completions path. DeepChat appends the path itself, so the URL should stop at /v1.
- Set the API URL to https://api.aicodewith.ai/v1 only β do not add /chat/completions
- API type must be OpenAI Completions, not Anthropic
- Do not leave a trailing slash
Model not enabled / model not found
Symptom: The chat footer cannot select the model you just added, or sending a message says the model does not exist.
Cause: The matching toggle was left off after connecting, or the Model ID is not in the list.
- Search the model list for a Model ID from this page and turn the toggle on
- If it is missing, click Add model and paste the ID
- Confirm the key can access that model
Connection fails / no chat reply
Symptom: Connect and load models fails, or sending never streams a reply.
Cause: The network is restricted, the provider is disabled, or the model is unavailable for this key.
- Confirm the custom provider is saved and enabled
- Try another model listed on this page
- Check AI Code With usage records for a matching call
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