FAQ
Common questions about service usage, the base URL, available models, endpoint support, prices and balance, streaming, function calling and leaked keys.
What can AI Nexus be used for?
This service is provided for learning and research purposes only. Please follow each model provider's usage policies.
What base URL should I use?
- OpenAI format (OpenAI SDK, most OpenAI-compatible clients):
https://ai.myaiexp.com/v1 - Anthropic format (Anthropic SDK, Claude Code):
https://ai.myaiexp.com, without/v1
Which models can I use?
Models are added over time; the current list and prices are on the model square, or call GET /v1/models with your API key.
Which endpoints support which models?
Endpoint support varies by model: /v1/chat/completions is the general OpenAI-compatible endpoint; /v1/messages works with models that support the Claude-native format; /v1/responses works with models that support the Responses API. The available models and the endpoints each one supports are listed on the model square.
See OpenAI SDK, Claude-native API and the Responses API.
Which models work with Claude Code and Codex CLI?
- Claude Code uses
/v1/messages; pick a model from the model square that supports the Claude-native format. - Codex CLI uses the Responses API; pick a model from the model square that supports the Responses API.
Why do I get "no available channel" (model_not_found)?
Usually the model ID is misspelled or has the wrong case, or you requested a model this service doesn't offer. Check the exact model ID with GET /v1/models first. See Errors.
How are prices set?
Model prices are synced daily from each vendor's official public list price, in USD per 1M tokens, and you are billed for actual usage. Current prices for every model are on the model square.
How do I check my balance and top up?
The console shows your balance in USD. All your keys share the account balance, and each key can also carry its own quota cap. Open the Wallet page to see your balance, top up or redeem a code, using the payment methods shown on the wallet page. See Wallet.
The answer is empty but there is no error?
Most likely a thinking (reasoning) model used up max_tokens while thinking. Increase max_tokens, or use a model without thinking.
Is the Responses API supported?
Yes, for models that support the Responses API (see the model square); for other models, use Chat Completions or the Claude-native Messages API. See Responses API.
Is image input supported? Image generation?
Image input (image understanding) is supported; see Multimodal. Image, audio and video generation are outside the scope of these docs.
Are streaming and function calling supported?
Yes; exact capabilities depend on the model you choose. See Streaming and Function calling.
Is prompt caching supported?
When calling a model that supports the Claude-native format through /v1/messages, enable it with cache_control; cache writes and reads are billed at the vendor's list price. See Prompt caching.
What if my key leaks?
Immediately delete or disable the key on the API Keys page, create a new one to replace it, and check Usage Logs for unexpected calls.
Where can I see call history and spending?
The console's Usage Logs page lists each request with its time, model, token counts and charge. See Usage Logs.
How do I sign up?
Sign up at https://ai.myaiexp.com/sign-up, then create a key on the API Keys page after signing in. See Quickstart for the full steps.
This service is provided for learning and research purposes only. Please follow each model provider's usage policies.