Run the EveryAPI gateway inside Aider
Aider is a terminal-based AI pair programmer that works inside a Git repository. Configure its OpenAI-compatible connection for EveryAPI and use an openai/ model name that is available in your account.
One base URL line points Aider at EveryAPI
Default global Base URL / China-first Base URL: https://api.everyapi.ai/v1 / https://api-cn.everyapi.ai/v1
# Install aiderpip install aider-installaider-install # Go through EveryAPI; the "openai/" prefix on --model tells aider to use the OpenAI-compatible protocolaider --openai-api-base https://api.everyapi.ai/v1 \ --openai-api-key sk-everyapi-... \ --model openai/claude-sonnet-5 # Aider then enters interactive mode — tell it "add type annotations to api.py"From zero to your first call
- Install aider
pip install aider-install && aider-install. The installed aider command is now in your terminal.
aider.chat/docs/install.html
- Configure the EveryAPI base URL + key
We recommend writing ~/.aider.conf.yml (persistent), but you can also pass --openai-api-base + --openai-api-key on the command line; env variables work too.
- Prefix the model name with openai/
Aider uses LiteLLM under the hood, and model names for an OpenAI-compatible endpoint must carry the openai/ prefix (e.g. openai/claude-sonnet-5). Otherwise LiteLLM will try to find the native Anthropic provider.
Aider × EveryAPI
Why does the model name need the openai/ prefix?
Aider uses LiteLLM under the hood to route providers. LiteLLM looks at the model-name prefix to decide which provider to use; openai/xxx goes through OPENAI_API_BASE / OPENAI_API_KEY. Without the prefix, LiteLLM treats claude-sonnet-5 as a request for the native Anthropic endpoint — a different path.