A self-hosted, Anthropic-compatible API for Claude models. Send the same requests your existing tools already make, and keep every call inside your own infrastructure.
Checking status…
Polling… · auto-refreshes every 15s
Drop-in Anthropic API
Point any Claude SDK or client at this base URL. The Messages API shape matches, including tool use and content blocks.
Streaming over SSE
Token-by-token responses stream as server-sent events, with per-app concurrency limits and clean back-pressure.
Private and observed
Runs on your own infrastructure. Every request is rate-limited, structured-logged, and recorded to the audit trail.
/v1/messages/v1/models/v1/account/health/health/deep/auditcurl https://your-gateway.example/v1/messages \
-H "x-api-key: $GATEWAY_KEY" \
-H "anthropic-version: 2023-06-01" \
-H "content-type: application/json" \
-d '{
"model": "sonnet",
"max_tokens": 1024,
"messages": [{ "role": "user", "content": "Hello, Claude" }]
}'© 2026 Private Claude-Compatible AI Gateway. Private deployment.