DeepSeek Models Now Available on AI Gateway via Azure
Key point
Vercel's AI Gateway has started supporting DeepSeek V4 Pro and Flash models via Azure.
Details
Vercel's AI Gateway has started supporting DeepSeek V4 Pro and V4 Flash models via Azure. This allows developers to use Azure as a failover path alongside their existing model providers to improve service reliability.
Azure is automatically included in the default routing without any code changes, and you can also configure Azure to be prioritized for specific models. If you're using the AI SDK, you can specify the call order via the order option in providerOptions.
AI Gateway provides the following core features:
- A unified API for model calls
- Usage and cost tracking, retry and failover configuration
- Custom Reporting and Zero Data Retention support
- Per-API-key budget settings and BYOK (Bring Your Own Key) support
Notably, AI Gateway passes through provider pricing as-is and does not charge any additional platform fee for inference, including for BYOK requests.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.