Commit History
-
fix: work on laya 0.3.2 too (PyPI index had not propagated 0.3.3)
Verified signature72ce472 -
perf: preload routing at boot, sharing the checkpoint the Space already builds
Verified signature8f9ff99 -
fix: import spaces before torch so ZeroGPU can attach a GPU
Verified signaturef3c3d8b -
fix: fall back to CPU when ZeroGPU cannot attach a GPU
Verified signature78a9a2d -
fix: routing tab runs on CPU, no background thread, no GPU fork boundary
Verified signaturece4ec65 -
fix: warm checkpoints on a background thread so start-up is never blocked
Verified signaturec40d817 -
perf: preload all checkpoints at boot so routing never pays a model load
Verified signaturec9da809 -
feat: multilingual routing tab (additive, existing tabs untouched)
Verified signaturee933e78
-
Remove credit card example from guardrails tab
Verified signature6c538b5 -
Rename space branding to Laya Demo
Verified signature07c37ab -
One shared GPU entry point, batched RAG scoring, per-call timing in logs
Verified signaturef7ccef9 -
Preload and warm the model at start-up so the first request is fast
Verified signature5b3b9ed -
RL Agent demo: triage, email, guardrails, RAG filtering, moderation, routing, playground
Verified signature938171e