Vercel AI Gateway now routes DeepSeek V4 Pro and V4 Flash through Azure as an additional provider, giving those two models a new failover path with zero code changes required.
Default routing includes Azure automatically. If a provider goes down, the gateway falls back through the remaining list. Developers who want Azure prioritized can set the order field in the gateway provider options for deepseek/deepseek-v4-pro or deepseek/deepseek-v4-flash inside the AI SDK. Existing Azure credentials work via BYOK, and AI Gateway passes through provider pricing with no markup and no platform fee on inference.
The full changelog is worth reading for the specific configuration syntax on provider ordering and BYOK setup, plus details on how AI Gateway handles retries, custom reporting, Zero Data Retention, and per-key budget controls. These are the operational knobs that matter when running DeepSeek at scale in production.
[READ ORIGINAL →]