Switchyard

Switchyard is an open-source model router that switches between multiple models to optimize the cost of requests. You can pass in two choices in the `models` parameter as well as specify any of the routing algorithms Switchyard offers. Details in the Switchyard docs.

ChatNVIDIA1M tokens-$1,090,059.6937 / -$1,090,059.6937 · 1M

nvidia/switchyard

Context
1M tokens
Completion cap
—
Tools
No
JSON
No
Released
2026-09-21

Call it from Geek Hub

Same OpenAI SDK. Change the base URL and the model id.

curl https://api.geekhub.mx/v1/chat/completions \
  -H "Authorization: Bearer $GEEKHUB_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nvidia/switchyard",
    "messages": [{"role": "user", "content": "Hola"}]
  }'
Get an API key

Frequently asked questions

What is Switchyard?
Switchyard is an open-source model router that switches between multiple models to optimize the cost of requests. You can pass in two choices in the `models` parameter as well as specify any of the routing algorithms Switchyard offers. Details in the Switchyard docs. Switchyard runs on the Geek Hub API (OpenAI-compatible). Model id: nvidia/switchyard.
Is Switchyard free?
No. Input is -$1,090,059.6937 per 1M tokens and output is -$1,090,059.6937 per 1M tokens on Geek Hub (markup included).
What is the context length of Switchyard?
Switchyard has a 1M tokens context window.
Does Switchyard support tool calling and structured outputs?
Switchyard does not advertise tool calling on this endpoint. Structured outputs are not listed for this endpoint.
When was Switchyard released?
Switchyard was released on 2026-09-21.

More models from NVIDIA