Aider

Aider connects to any OpenAI-compatible endpoint via a single environment variable — no plugin, no config file required (though both are supported too). Point that variable at Tempr Gateway and every model on your allowlist is one --model flag away, under one key, instead of juggling a separate API key per provider.

1. Create a virtual key

   In the Portal, create a Gateway virtual key. Keys are prefixed tvk_ and shown once at creation.

2. Add a provider key

   Gateway is BYOK: add a key for at least one provider from the supported list.

3. Set the environment variables

   Before running aider:

# Mac/Linux
export OPENAI_API_BASE="https://api.temprhq.io/v1"
export OPENAI_API_KEY="tvk_..."

# then run aider with a model on your allowlist
aider --model openai/anthropic/claude-sonnet-4-5

Aider expects OpenAI-compatible models to be prefixed with openai/ in its own --model flag (this is Aider's own provider-routing prefix, on top of Tempr's own provider/model id) — the example above requests Tempr's anthropic/claude-sonnet-4-5 through that openai/ passthrough.

What's next

• Authentication & virtual keys — scoping keys with model allowlists, budgets, and rate limits.
• Models, limits & logs — request logs and analytics for everything you send through the gateway.