Non-autoregressive System 1 decision engine. Typed choice, score and yes/no decisions over any text in a single forward pass, in 100+ languages, with a router that picks the right checkpoint per request.
Python calibrationclassificationdecision-modelhuggingfacejevmodernbertmultilingualnlppythonpytorchroutingtyped-decisionszero-shot
Commits on main
laya-go: hooks (OnPredictStart/End/Error), registries, dispatch with timeout and panic recovery
b36c8a5 Adeshina Lasisi committed 1 month ago
test(evals): pin that questions_fingerprint hashes a sorted set of schemas
5ed6daa aashish254 committed 1 month ago
fix(mcp): refuse a LAYA_DEVICE index torch rewrites instead of building another GPU
6f0a355 Aashish committed 1 month ago
laya-go: pin that a scanned PredictLong writes no gate field
dbb12df Adeshina Lasisi committed 1 month ago
laya-go: document min_confidence on PredictBatch; raise the test floor to 1447
4087bb0 Adeshina Lasisi committed 1 month ago
fix(hooks): refuse a bool hooks_timeout instead of reading it as 1.0
ae132f4 Aashish committed 1 month ago
fix(backends): read LAYA_INDUCTOR_CACHE_DIR stripped, so a padded path is the path
b0bb484 aashish254 committed 1 month ago
fix(evals): escape Markdown cell content in the report's slice labels
9dc4f7e Aashish committed 1 month ago
test(runtime-fixes): stage markers, so the Windows crash names its section
083a229 Nandakishor committed 1 month ago
docs(examples): the marker guard no longer re-renders the options
37d7a36 Nandakishor committed 1 month ago