SDKs

Nemo Router speaks the OpenAI REST shape, so any OpenAI-compatible client works without a custom SDK. Point the client at https://api.nemorouter.ai/v1, pass your sk-nemo-... key, and pick a model — guardrails, prompt templates, caching, rate-limits, and cost tracking all run server-side.

Pick your language

The only three things every SDK needs

1. Base URL — https://api.nemorouter.ai/v1
2. API key — your sk-nemo-... virtual key, sent as Authorization: Bearer sk-nemo-...
3. Model — any model string from the models catalog

curl https://api.nemorouter.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-nemo-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-2.5-flash",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

What you get for free

Without changing a line of SDK code:

• Guardrails — PII redaction, prompt injection blocking, keyword filters
• Caching — repeat prompts return instantly from cache
• Prompt templates — versioned, A/B-tested, injected server-side
• Rate limits — per-key / per-org RPM/TPM enforcement
• Cost tracking — every response carries x-nemo-request-cost
• Model fallbacks — switch providers by changing the model name string

Using a framework instead?

If you build with LangChain, LlamaIndex, Vercel AI, CrewAI, AutoGen, or Google ADK, see the Frameworks guides — same base-URL swap, no plugin required.