ToolsSeptember 2, 2026via MarkTechPost
Meet Switchyard: A Rust Proxy and Library That Routes and Translates LLM Traffic Across OpenAI and Anthropic APIs
Why it matters
Switchyard lowers switching costs between LLM providers by abstracting API differences into a single routing layer. Practitioners building multi-model or multi-vendor strategies gain a concrete option for provider agility without app rewrites.
Key signals
- NVIDIA released Switchyard under Apache 2.0 license
- Rust-based proxy and library for LLM traffic translation
- Supports OpenAI, Anthropic APIs, vLLM, NIM, Ollama as targets
- Routes via passthrough, random, LLM-classifier, or stage-router algorithms
- Pre-alpha, not production-ready
- Decodes requests into provider-neutral types, translates responses back to client format
- Enables tools like Claude Code or Codex CLI to run against different backends unchanged
- NVIDIA released Switchyard as Apache-2.0 open-source Rust library/proxy
- Supports provider-neutral request/response translation between OpenAI and Anthropic APIs
- Routing modes: passthrough, random, LLM-classifier, stage-router
- Enables tools like Claude Code and Codex CLI to run against vLLM, NIM, or Ollama backends unchanged
- Abstracts API differences for client applications
The hook
NVIDIA open-sources Switchyard: a Rust proxy that lets you swap LLM providers (OpenAI, Anthropic, vLLM, Ollama) without rewriting client code.
NVIDIA has released Switchyard, an Apache-2.0 Rust proxy and library for LLM traffic. It decodes requests into provider-neutral types, routes them with passthrough, random, LLM-classifier, or stage-router algorithms, and translates responses back into the client's format, so Claude Code or Codex CLI…