Demo mode — sample data. Editing is disabled.

Switchboard

Assign each API key a fixed model (plus fallbacks) that requests using mizan/switchboard resolve to — and switch it anytime with no redeploy. A key without a configuration falls back to mizan/auto.

How it works

Use mizan/switchboard as the model in your code. The gateway resolves it per API key — no hard-coded models, no redeploys.

Swap the primary model (or its fallbacks) here any time. Changes apply to the key within seconds.

Choose Mizan billing (your wallet) or BYOK billing (your own provider account) per key.

Production

active

mizan_rt_demo0xxxxxxxxxx… · Production

Primary model: anthropic/claude-sonnet-4-5

Fallback models: openai/gpt-4o, gemini/gemini-2.5-flash

Billing: Mizan billing

Requests

125

Spend

$0.86

Fallback rate

100%

Served by (last 30 days)

mistral/mistral-large-latest22 · $0.13

grok/grok-2-latest19 · $0.15

openai/gpt-4o17 · $0.19

anthropic/claude-sonnet-4-517 · $0.30

groq/llama-3.3-70b-versatile17 · $0.03

deepseek/deepseek-chat17 · $0.02

gemini/gemini-2.5-flash16 · $0.04

Staging

active

mizan_rt_demo1xxxxxxxxxx… · Internal Tools

Not configured — mizan/switchboard requests behave like mizan/auto.

No Switchboard traffic in the last 30 days.

Mobile App

active

mizan_rt_demo2xxxxxxxxxx… · Default workspace

Not configured — mizan/switchboard requests behave like mizan/auto.

No Switchboard traffic in the last 30 days.

Analytics

active

mizan_rt_demo3xxxxxxxxxx… · Data Pipeline

Not configured — mizan/switchboard requests behave like mizan/auto.

No Switchboard traffic in the last 30 days.