Commit Graph

1 Commits

Author SHA1 Message Date
bvandeusen 52d6a8ed53 feat(embeddings): swap Ollama for fastembed (in-process ONNX)
Replaces the Ollama HTTP get_embedding with a fastembed.TextEmbedding
singleton loaded lazily on first call. Model: BAAI/bge-small-en-v1.5
(384-dim), cached to /data/fastembed-cache.

Public API unchanged:
  - get_embedding(text, model=None) — `model` now silently ignored
  - upsert_note_embedding
  - semantic_search_notes
  - backfill_note_embeddings

_cosine_similarity gains a defensive length-mismatch check so any
stale 768-dim row that survived the migration is treated as 0.0
similarity rather than crashing zip().

The Ollama client dep stays in pyproject for now (other services still
use it); Phase 7 removes it once chat/journal/curator are gone.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-26 21:02:30 -04:00