openai
// nvidia
Switchyard
nvidia/switchyard
Switchyard has no published list price with a 1M-token context window.
A model from nvidia, released Sep 21, 2026. LiveBench hasn't published a run for it, so this page carries measured price and specs only — no benchmark numbers.
- Input / 1M
- —
- Output / 1M
- —
- Context
- 1M
- LiveBench overall
- Not evaluated
Specs and pricing
| Input price USD per 1M prompt tokens | — |
|---|---|
| Output price USD per 1M completion tokens, reasoning included | — |
| Blended price 3:1 input:output | — |
| Cached input read Blank means not priced separately — not free | — |
| Cache write | — |
| Context window | 1,000,000 tokens |
| Max output | — |
| Input modalities | text |
| Tool calling | No |
| Extended reasoning | No |
| Knowledge cutoff | — |
| Listed | Sep 21, 2026 |
Compare Switchyard with
Benchmarked models at a similar price.
anthropic
Switchyard vs Claude Sonnet 5.5
openai
Switchyard vs GPT-6 Luna
openai
Switchyard vs GPT-6 Sol
anthropic
Switchyard vs Claude Opus 5.5
x-ai
Switchyard vs SpaceXAI: Grok 4.7
Switchyard FAQ
What is the context window of Switchyard?
Switchyard accepts up to 1,000,000 tokens of context.
Does Switchyard support tool calling?
Switchyard does not list tool (function) calling support. It accepts text input.
How does Switchyard score on benchmarks?
LiveBench has not published a run for Switchyard yet. Until it does, compare it on price and specs, or run your own evaluation on your own tasks.
What is the API model ID for Switchyard?
On OpenRouter the model ID is "nvidia/switchyard". It was listed on Sep 21, 2026.
More from nvidia
How these numbers are produced
- Price and specs — provider list data from OpenRouter, refreshed every 15 minutes. “Blended” is a 3:1 input:output mix.
- Scores and ranks — LiveBench release 2026-06-25. Ranks count only models that have a published run; a blank means “not evaluated”, never “bad”.
- Cost per point — the measured dollars LiveBench spent on the run, divided by the score it earned.
Published benchmarks rank models on someone else's tasks. Before committing, see LLM & agent evaluation for building an eval on your own.