Zed

Zed's Agent Panel supports custom OpenAI-compatible providers, configured either through the UI or directly in settings.json. Add Tempr as one and list every model on your allowlist under it — Claude, GPT, Gemini — all switchable per-conversation from the same model dropdown, under one key.

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. Add Tempr as a provider

   Open the Agent Panel (Cmd/Ctrl+Shift+A) → + Add Provider, or edit ~/.config/zed/settings.json directly:

{
  "language_models": {
    "openai_compatible": {
      "tempr": {
        "api_url": "https://api.temprhq.io/v1",
        "available_models": [
          {
            "name": "anthropic/claude-sonnet-4-5",
            "display_name": "Claude Sonnet 4.5 (via Tempr)",
            "max_tokens": 200000
          }
        ]
      }
    }
  }
}

Zed derives the API key environment variable from the provider id you chose (tempr above) as upper snake case plus _API_KEY — so set TEMPR_API_KEY to your virtual key before launching Zed, or enter it through the Agent Panel's settings UI, which stores it in your system keychain instead.

Model ids

Use the fully-qualified provider/model form (e.g. anthropic/claude-sonnet-4-5, openai/gpt-5) for each entry in available_models — the same ids GET /v1/models returns.

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.