Vercel adds more ZDR providers for Kimi K3.1 Pro on AI Gateway

Kimi K3.1 Pro and Kimi K3.1 Pro Highspeed are now available through more Zero Data Retention providers on Vercel’s AI Gateway. The change gives users more routing options for models that support image input and a 1M-token context window.
- Kimi K3.1 Pro and Highspeed now have more ZDR providers on AI Gateway.
- Both models offer a 1M-token context window and image input.
- ZDR can be required for individual requests or an entire team.
- Requests fail if no ZDR provider is available for the selected model.
- AI Gateway supports coding agents, routing, usage tracking, and failover.
More provider choice for Kimi models
Until this update, Moonshot AI was the only provider serving Kimi K3.1 Pro and Kimi K3.1 Pro Highspeed on AI Gateway. Vercel says both models can now be accessed from more providers with Zero Data Retention, or ZDR, agreements.
Kimi K3.1 Pro is positioned for coding, 3D modeling, and financial analysis. Vercel also says it can support long agent runs, from a small fix through to a full project. The Highspeed version is designed for interactive work such as coding agents, where response time matters.
How developers can require ZDR
Developers can call Kimi K3.1 Pro through the AI SDK, the Chat Completions, Responses, and Anthropic Messages APIs, or a coding agent connected to AI Gateway. The model name is moonshotai/kimi-k3.1-pro. For the faster version, users should use moonshotai/kimi-k3.1-pro-highspeed.
At the request level, setting zeroDataRetention in the gateway provider options requires ZDR for that request. Teams can also turn on ZDR in the AI Gateway dashboard settings to apply the requirement to every request, including requests from coding agents.
When either setting is enabled, AI Gateway routes requests only to providers with a ZDR agreement. If no such provider is available for the selected model, the request fails instead of being sent to a provider without one.
Support for coding agents and model operations
Vercel says Kimi models can be used with Claude Code, Codex, Cursor, and other coding agents. To configure them, users need the latest Vercel CLI, then run the coding agents setup command and select one of the two Kimi model names.
AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations for higher-than-provider uptime. It also includes custom reporting, ZDR support, API key budgets, and routing rules.
According to Vercel, AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including Bring Your Own Key requests. Developers can also try Kimi K3.1 Pro in the model playground.
Vercel: Kimi K3.1 Pro now has more ZDR providers on AI Gateway


