⚡NemoClaw + Tokaroo integration
Tokaroo + NemoClaw
One API key. Real-time AI routing across 199+ AI models from Anthropic, Google, Groq, and OpenAI - chat, images, speech, and video - optimized for cost, latency, and automatic failover. Connect in under 60 seconds.
Set up in 4 steps
01
Get your API key
Sign up at tokaroo.com - it takes 30 seconds. Your key starts with tok_.
Get API key ->02
Add Tokaroo to your NemoClaw provider config
Add the tokaroo provider block to your NemoClaw configuration. NemoClaw supports the same OpenAI-compatible provider format as OpenClaw.
// nemo_config.json (or your NemoClaw provider config)
{
models: {
mode: "merge",
providers: {
tokaroo: {
baseUrl: "https://api.tokaroo.com/v1",
apiKey: "${TOKAROO_API_KEY}",
api: "openai-completions",
models: [
{ id: "auto", name: "Tokaroo Auto - best value, automatic routing" },
{ id: "fast", name: "Tokaroo Fast - speed-optimized, low latency" },
{ id: "max", name: "Tokaroo Max - highest capability" },
]
}
}
},
agents: {
defaults: { model: "tokaroo/auto" }
}
}03
Set your API key
Add it to your environment or NemoClaw's .env file.
# Set in your environment or NemoClaw's .env file
TOKAROO_API_KEY=tok_...04
Restart NemoClaw
Every LLM call in your NemoClaw agents now routes through Tokaroo - automatic routing, semantic cache, and auto fallback.
What your agents get
Enterprise-ready routing
NemoClaw's OpenShell gateway routes through Tokaroo's optimization engine. Every request gets the best model at the lowest cost - automatically.
Semantic cache
Repeated or similar queries answered from cache. In enterprise workloads with high prompt repetition, cache savings compound quickly.
Auto fallback
If any upstream provider goes down, Tokaroo re-routes silently. No agent crashes, no manual intervention.
Usage analytics
Per-request cost and latency visible in the Tokaroo dashboard. Full spend history with no blind spots.
OpenAI-compatible
Tokaroo speaks the same API as OpenAI. No custom integrations, no special SDK. If NemoClaw can talk to OpenAI, it can talk to Tokaroo.
On-prem + cloud hybrid
Running NemoClaw on-prem with vLLM or NVIDIA NIMs? Tokaroo routes between your on-prem inference and cloud providers automatically.
About NemoClaw
NemoClaw is NVIDIA's enterprise fork of OpenClaw, featuring the OpenShell gateway and deep integration with NVIDIA NIM inference microservices. It uses the same OpenAI-compatible provider configuration as OpenClaw, so the Tokaroo integration works identically across both. If you're running NemoClaw with on-prem NIM endpoints, Tokaroo can route between your local inference and cloud providers automatically.
Also see
Ready to connect?
Create your API key and have your NemoClaw agents routing through Tokaroo in under a minute.
Create API key