# NVIDIA-NeMo/Switchyard

> An LLM request router that picks the cheapest model capable of handling each call while preserving OpenAI and Anthropic API compatibility. Runs as a NeMo Relay or LiteLLM plugin, an embeddable library, or a standalone proxy.

- Magnitude: 1.4 out of 10 — Cooling
- Stars: 3,207 total · +3 stars on 2026-09-23 (UTC), ≈ 8 by evening
- Star trust: star growth looks organic
- Category: Language models · Language: Rust · License: Apache-2.0 · Created: 2026-05-19 · Last push: 2026-09-23
- GitHub: https://github.com/NVIDIA-NeMo/Switchyard · Page: https://gitnova.dev/en/r/NVIDIA-NeMo/Switchyard

## Useful for

- Attach a routes.toml to an existing LiteLLM Router to cut token costs
- Run the standalone proxy and point Claude Code or Codex CLI at it without code changes
- Embed the routing algorithm into your own gateway via pip install nemo-switchyard

## Why it’s here

- 3 stars so far today, about 8 expected by the end of the day.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 304
- Issues and pull requests: 828
- Watchers: 10
- Average over the last week: 11 per day
- Usual pace: 43 per day
- Latest release: v0.3.0 (2026-09-22)

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-09-30 … 2026-10-29: None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None

## Magnitude by day, last 11 days

09-13 3.1, 09-14 3.0, 09-15 2.2, 09-16 2.1, 09-17 2.1, 09-18 2.0, 09-19 1.8, 09-20 1.6, 09-21 1.4, 09-22 1.4, 09-23 1.4

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-09-23 10:35 UTC, updated every 30 minutes.