Expand description
Inference invoke seam — exposes semantic skills, grounding, and post-turn verification.
Future crate: qualia-inference.
Functions§
- conditioning_
activate Conditioning.activate— move the active registry pointer after approval gates.- conditioning_
compile Conditioning.compile— compile against an explicit backend capability and authority snapshot.- conditioning_
evaluate Conditioning.evaluate— start a bounded evaluation campaign.- conditioning_
inspect Conditioning.inspect— return a redacted requirement/token/evidence trace.- conditioning_
rollback Conditioning.rollback— restore a prior approved version.- conditioning_
validate Conditioning.validate— validate one profile and return structured violations.- constrained_
decode Inference.constrained_decode— run DOMINO constrained decoding on a token sequence.- detect_
ungrounded Inference.detect_ungrounded— check if a generation output is ungrounded. Returns a boolean and optional reason.- embed
Inference.embed— embed text into a vector using the default TextEmbedder.- gelu
Inference.gelu—{ x: [f64] }→{ out: [f64] }.- grounding
Inference.grounding— check whether a generation is grounded against the quant graph fact store. Returns a record withtext,repaired, and optionalreason.- load_
model Inference.load_model— load a GGUF/P64 model into resident memory.- relu
Inference.relu—{ x: [f64] }→{ out: [f64] }.- rms_
norm Inference.rms_norm—{ x, weight: [f64], eps? }→{ out: [f64] }.- run_
classifier Inference.run_classifier— run a statistical classifier (KNN, NB, or SVM).- run_
reranker Inference.run_reranker— rerank candidate documents by relevance to a query.- run_
transformer Inference.run_transformer— run a transformer forward pass.- sigmoid
Inference.sigmoid—{ x: [f64] }→{ out: [f64] }.- softmax
Inference.softmax—{ x: [f64] }→{ out: [f64] }(sums to 1).- unload_
model Inference.unload_model— drop the resident model mmap. Native only — returns E300 on WASM.- vector_
search Inference.vector_search— search the vector store for nearest neighbours.- verify_
turn Inference.verify_turn— verify and heal a completed generation turn. Returns the final text, repair status, and individual checks.