AI Transit Narrative (single date)

Bring your own model key: the LLM provider that should run this completion. Send together with X-Provider-Key or the request is refused: a key with no provider bills us for tokens you meant to pay for, and a provider with no key would look like it worked. The endpoint is fixed per provider on our side and is never taken from the request. There is no fallback to our own chain: if your key fails you get the provider's status and message.

Your own credential for X-Provider. Never logged, stored or cached: it travels in a header because the request body is hashed into two cache keys, and a BYOK turn bypasses both caches in each direction. A provider error that quotes the key back is redacted before it reaches our logs or your response. Such a turn costs 5 credits instead of the endpoint price, and only when a completion actually ran on your key.

Optional model id, overriding the default for that provider. Defaults: openai → gpt-4o-mini, anthropic → claude-sonnet-4-5, google → gemini-2.5-flash, groq → llama-3.3-70b-versatile, mistral → mistral-large-latest, openrouter → meta-llama/llama-3.3-70b-instruct, deepseek → deepseek-chat, together → meta-llama/Llama-3.3-70B-Instruct-Turbo.