Skip to content

Switchyard lets LLM applications route traffic across models and providers while preserving native OpenAI and Anthropic API compatibility - enabling flexible model selection, benchmarking, and cost/performance optimization.

Rust

Commits on main

build(python): support Python 3.10 (#391)
nachiketb-nvidia committed 3 months ago
0151621
feat: Introduce ModelId type (#373)
Graham King committed 3 months ago
48b3b71
2bef154
refactor(python): remove deprecated server stack (#343)
nachiketb-nvidia committed 3 months ago
b104765
feat(server): expose stage router stats (#344)
nachiketb-nvidia committed 3 months ago
6df31a9
fb3fc30
feat(libsy): emit stage router metrics directly (#342)
nachiketb-nvidia committed 3 months ago
bced06f
docs(release): finalize 0.2.0 changelog (#339)
nachiketb-nvidia committed 3 months ago
d7911fe
chore: Rename `CallLlm` to `CallModel` (#340)
Graham King committed 3 months ago
94872ce
refactor(libsy): make Decision a struct (#338)
Ayush Agarwal committed 3 months ago
4fc554a
feat: Remove TypeErasedDriver (#337)
Graham King committed 3 months ago
3948b4c
fix(release): make package publishing resumable (#336)
nachiketb-nvidia committed 3 months ago
c19f3d0
refactor(python): remove legacy routing profiles (#268)
nachiketb-nvidia committed 3 months ago
f30498d
e8390d4
ci(release): publish Rust crates from release tags (#329)
nachiketb-nvidia committed 3 months ago
5111678
8d6adfe
74b8990
docs: add 0.2.0 known issues (#327)
ryan-lempka committed 3 months ago
9d7cea4
fix(server): return Codex metadata from /v1/models (#304)
Elyas Mehtabuddin committed 3 months ago
759658e