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
refactor(python): remove dead chain processors (#127)
7315625 nachiketb-nvidia committed 3 months ago
fix(container): make server image non-root compatible (#128)
3bd5fa6 Bhuvan Agrawal committed 3 months ago
fix(launch): guard zero-config default trio and report routing bundle in --dry-run (#108)
d4c37dc Elyas Mehtabuddin committed 3 months ago
docs: correct AUTO probe order, extras, saved-bundle note, and stale API references (#110)
b28bcf6 Elyas Mehtabuddin committed 3 months ago
fix(stats): record noop routes in /v1/routing/stats (#109)
8baf8ec Elyas Mehtabuddin committed 3 months ago
feat(serve): route-log per-request routing stats for profile-config serve (#118)
d0a924c Brian Newsom committed 3 months ago
feat(libsy): add OTel metrics, tracing spans, and structured logging (#69)
5059359 eric-liu-nvidia committed 3 months ago
refactor(profiles): nuke components-v2 and its mixed Python serving stack (#119)
4ccec5d nachiketb-nvidia committed 3 months ago
feat(profiles): route sub-agent requests to a profile-level subagent_target (#112)
060ad75 Lin Jia committed 4 months ago
feat(benchmark): net out retried attempts by trial id in task-level routing stats (#120)
ee5e96d ryan-lempka committed 4 months ago
feat(libsy): stateful FallThrough routing algorithm + composable building blocks for libsy (#101)
2b62048 Greg Clark committed 4 months ago
docs(local-models) [DO NOT MERGE]: add vllm hidden-state serving notes (#13)
832547f sabhatinas committed 4 months ago
fix(translation): forward prompt-cache token usage across format translation (#117)
097d0e4 Brian Newsom committed 4 months ago
feat(server): load libsy algorithms from explicit TOML (#115)
f47ab9b nachiketb-nvidia committed 4 months ago
feat(benchmark): task-level routing stats for Harbor runs (#116)
f46703b ryan-lempka committed 4 months ago
fix(libsy): propagate errors and cancel the algorithm task in `run_stream` (#114)
f91b61a Greg Clark committed 4 months ago
refactor(server): serve libsy algorithms without profiles (#98)
e09b5b0 nachiketb-nvidia committed 4 months ago
feat(libsy-protocol): add Metadata::from_headers harness header normalization (#92)
89edae4 Greg Clark committed 4 months ago
chore: Move libsy-examples ensemble to libsy/examples (#102)
2fa6b87 Graham King committed 4 months ago
feat(intake): log each router's own model call (#74)
91e77df Elyas Mehtabuddin committed 4 months ago
fix(translation): attach responses reasoning items to their assistant turn instead of emitting empty chat messages (#99)
5ae2a0f eric-liu-nvidia committed 4 months ago
feat(switchyard-llm-client): add HTTP LLM client crate (#79)
0bc032a Greg Clark committed 4 months ago
fix(cli): make configure inputs a typed request so first-run setup can't crash (#94)
932a938 Elyas Mehtabuddin committed 4 months ago
docs(configure): require api key in route-bundle setup (#30)
88cbc24 sabhatinas committed 4 months ago