Switchyard API: toman pricing and code
nvidia/switchyard
Not enough requests yet for an average
This model is a router: it sends each request to the model it judges best for it, and the request costs that model's price. Before sending, an amount is set aside from your wallet; after the answer only the real cost is deducted and the rest returns at once. The exact cost of every request is in your dashboard's usage history.Pricing and top-ups
Switchyard example: summarising a document
The example is picked from this model's own capabilities. Drop in your key and it runs as is.
from openai import OpenAI
client = OpenAI(base_url="https://api.uttapen.ir/v1", api_key="sk-up-…")
stream = client.chat.completions.create(
model="nvidia/switchyard",
messages=[{"role": "user", "content": "Summarise this text in three sentences:
\"This year's budget act makes three main changes…\""}],
stream=True,
)
for chunk in stream:
print(chunk.choices[0].delta.content or "", end="", flush=True)import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://api.uttapen.ir/v1", apiKey: "sk-up-…" });
const stream = await client.chat.completions.create({
model: "nvidia/switchyard",
messages: [{ role: "user", content: "Summarise this text in three sentences:
\"This year's budget act makes three main changes…\"" }],
stream: true,
});
for await (const chunk of stream) process.stdout.write(chunk.choices[0]?.delta?.content ?? "");curl https://api.uttapen.ir/v1/chat/completions \
-H "Authorization: Bearer sk-up-…" \
-H "Content-Type: application/json" \
-d '{
"model": "nvidia/switchyard",
"messages": [{"role": "user", "content": "Summarise this text in three sentences: \"This year's budget act makes three main changes…\""}],
"stream": true
}'What is Switchyard good for?
NVIDIA publishes this model; we expose it under the id "nvidia/
This one handles text only. If you need function calling or image understanding, pick another model from the same catalogue — swapping is a one-line change.
NVIDIA has 12 models in our catalogue; the cheapest is Nemotron 3.5 Lightning at 122 toman per thousand words and the dearest Nemotron 3 Ultra at 1,350. It does not support include_reasoning, max_tokens, reasoning, response_format, which most models here do, so test before switching if your code relies on them. By context size the nearest option from another provider is Nova 2 Lite at 1,000,000 tokens.
Good fits: automatic routing when you would rather not pick a model yourself, analysing a long document or codebase in one request.
Switchyard is an open-source model routing system that switches between multiple models to optimize the cost of requests. By default it will use this platform market data to select the most...
Frequently asked
- How do I call Switchyard from Iran?
- Sign up with your mobile number, top the wallet up in toman, create an API key, then in the official OpenAI SDK point base_url at https:/
/ api.uttapen.ir/ v1 and set model to "nvidia/ switchyard". Nothing else in your code changes. - What does Switchyard cost in toman?
- This model has no fixed price: the cost depends on the model that answered the request, and that exact amount is what leaves your wallet.
- How much input does Switchyard take?
- Up to 1,000,000 tokens per request, roughly 750k words. Cap the answer length yourself with max_tokens.
- Can I stream Switchyard's output?
- Yes — with stream=true you get SSE events as the tokens are produced. This model has no tool calling and no image input, so pick a different one if you need either.