Serverless

Pay-as-you-go access to Wafer models through OpenAI-compatible and Anthropic-compatible APIs.

Wafer Serverless gives you pay-as-you-go access to Wafer-hosted models through the same inference endpoint used across Wafer products. Load credits, mint an API key, and pay per token with no subscription window.

Connection Details

OpenAI-compatible endpoint	https://pass.wafer.ai/v1
Anthropic-compatible endpoint	https://pass.wafer.ai/v1/messages
Send your API key as	Authorization: Bearer <key>
Request-scoped ZDR	Add Wafer-ZDR: required on direct API calls

API Capabilities

Serverless supports request-scoped privacy and advanced completion controls:

• API Reference
• Zero Data Retention
• Tokenized Completions and Constrained Decoding
• Metrics API
• Usage API

Models

Call GET https://pass.wafer.ai/v1/models to see the currently public Serverless catalog. Each model card includes whether the model supports ZDR.

curl -sS "https://pass.wafer.ai/v1/models" \
  -H "Authorization: Bearer <YOUR_WAFER_API_KEY>"

The public catalog only lists models that are ready for general use. Some routed models may be available by allowlist before they appear in the public catalog.

Agent Setup

For Claude Code, Codex, Cline, Roo Code, and other harnesses, see Agent Setup.

Router Setup

For Vercel AI Gateway, OpenRouter, and TrueFoundry AI Gateway, see Router Setup.

For direct curl requests and request parameters, see API Reference.