LobeHub × AI Code With
Use AI Code With in LobeHub: Third-Party API / Relay Setup
This page shows how to connect a third-party API in open-source LobeHub so you can manage models and tasks with a custom provider.
Have a reachable LobeHub instance ready, then add the AI Code With provider in Settings; the API key is only entered in LobeHub, never on this page.
- Provider Name
- AI Code With
- Proxy Address (Base URL)
- https://api.aicodewith.ai
- API Key
- sk-xxxxxxxx
- Model
- deepseek-v4-flash
Quick setup
Add AI Code With as a third-party provider in LobeHub settings, and use DeepSeek, Gemini, GLM, Kimi, and more models for chat and task assignment in about 3 minutes.
LobeHub (formerly LobeChat) is an AI chat and agent platform that supports custom OpenAI-compatible providers. Add AI Code With once in Settings, and every model is available for chat and task assignment — no plugin and no config file needed.
Click + to add a custom provider
Open Settings → AI Providers. Click the plus next to Search providers to add a custom provider. Do not switch to built-in LobeHub Cloud.
Open Settings → AI Providers. Click the plus next to Search providers to add a custom provider. Do not switch to built-in LobeHub Cloud. Pick New API and fill the host plus key
Select New API. Paste your AI Code With API key (it stays on your device only — never share your real key in screenshots or chats). Set the API address to https://api.aicodewith.ai with no /v1 — the form hint says it usually should omit /v1. Then fetch the model list.
Quick copy
API address (no /v1)
https://api.aicodewith.ai
Model ID
deepseek-v4-flash
Pick New API. Paste your AI Code With API key. Set the API address to https://api.aicodewith.ai with no /v1 (the form hint says it usually should omit /v1). Then fetch the model list. Disable all, then enable the models you need
After fetching models, click Disable all in the top right, then enable the AI Code With models you need. Do not keep sending with built-in LobeHub models.
Copy a Model ID
After adding the provider, click "Get model list" to fetch models automatically, or type one of these Model IDs manually:
Pick a family, then copy a Model ID. You can fetch the list automatically, or paste it by hand.
Selected
- Provider
- DeepSeek
- Model ID
- deepseek-v4-flash
The full model list follows the AI Code With docs and pricing page; this site does not maintain a separate model list.
View the full model listAfter fetching models, click Disable all in the top right, then enable the AI Code With models you need. Do not keep sending with built-in LobeHub models. Assign custom models on Service Models and verify
Open Service Models and assign each task to a custom model you just enabled. A normal result confirms the connection — then match the call by time and model name in AI Code With usage records.
Open Service Models and assign each task to a custom model you just enabled. A normal result confirms the connection; then match the call by time and model in AI Code With usage records.
How does LobeHub connect to AI Code With?
LobeHub supports custom OpenAI-compatible providers. Connecting to AI Code With only needs 3 settings:
- New API
- In LobeHub Settings, click New API in the provider list to add an OpenAI-compatible provider (the exact label may vary by deployment version).
- API Key
- Paste your AI Code With API key. The key stays in your LobeHub instance and is never sent to any third-party server.
- API address
- Enter https://api.aicodewith.ai with no /v1. The LobeHub New API form hint says it usually should omit /v1.
Flow: LobeHub instance → custom OpenAI-compatible provider (New API) → AI Code With API. Everything is configured inside LobeHub, no files to edit (self-hosted setups can use environment variables).
Frequently asked questions
New to LobeHub?
LobeHub is an open-source AI agent and chat platform. Use it on the official cloud, or self-host it.
LobeHub official website
Visit the LobeHub website to learn about the cloud and self-hosted options. Cloud instances work right after login; self-hosting follows the official Docker guide.
Open the LobeHub websiteFull deployment and usage guidance follows the LobeHub official docs; this site does not host LobeHub images or installers.
Common error troubleshooting
The most frequent problems fall into 4 categories.
401 / Invalid API Key
Symptom: The request returns 401, Unauthorized, or Invalid API Key.
Cause: Usually the API key is invalid, mistyped, or copied with extra characters.
- Re-paste the API key in the LobeHub provider settings (no extra spaces)
- Confirm the key is a valid AI Code With API key
- Retry sending a message
Model Not Found / Wrong Model ID
Symptom: The request reports Model Not Found or an unknown model.
Cause: The Model ID is mistyped, or the model is currently unavailable.
- Check the Model ID in the provider settings matches the one on this page
- Confirm the key can access that model
- Confirm the model is available on the AI Code With pricing page
API address error
Symptom: The request fails on test or when sending a message, or the model list is empty.
Cause: The API address has a wrong path, or an extra /v1 was added.
- Confirm the value is https://api.aicodewith.ai (no /v1)
- Remove any trailing /v1, /v1/models, or guessed path
- Save the settings and test again
Empty model list
Symptom: Clicking "Get model list" returns no models.
Cause: The API address is unreachable, the network is restricted, or the endpoint does not support auto-fetching the model list.
- Confirm the API address is https://api.aicodewith.ai (no /v1)
- Switch network and retry
- Type a Model ID manually to use the provider
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