Overview

Connecting

Connecting to upstream

The gateway is up; traffic is served as soon as the first upstream request succeeds.

Active0
Queued0
Uptime0ms
Requests (24h)00 completed
Success rate—0 failed · 0 cancelled
P50 generation—0 cooldowns
Input tokens00 output
Cache share—0 cached
Connect
OpenAI-compatible base URL. Works with any client that accepts a custom endpoint.
Base URL
—
Use exactly this value, including /v1.
Model
deepseek-v4.1-flash
Any model id is served by this model; this is the canonical id.
Authentication
Authorization: Bearer <gateway key>
The upstream provider key never leaves the server.
Max request body: —Max output: 0 tokens · default 0
POST /v1/chat/completions
curl https://your-gateway.example.com/v1/chat/completions \
  -H "Authorization: Bearer $GATEWAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4.1-flash",
    "messages": [{"role": "user", "content": "Hello"}],
    "stream": true
  }'
Live queue0 active · 0 queued
Strict FIFO — exactly one upstream request is in flight at any time.
Queue is empty
New requests appear here the moment they arrive.
Modelactive
Served through the Anthropic Messages transport, which preserves image input and cache accounting.

Context

0

payload dependent

Max output

0

tokens per request

ReasoningTool callingVisionSSE streaming

Cached prompt tokens are reported per request. The gateway translates OpenAI requests to the provider dialect and back.

Recent requestsView log
Latest traffic through the queue.
Activity
Gateway lifecycle events.
No events yet
Traffic events stream in here.
Requests over time
Completed, failed and cancelled requests per bucket.
ErrorsAll requests
Failed requests in the last 24 hours.