feat(ml): default to SigLIP 2 (new installs) + model dropdown, no free-text (#1203)
CI / lint (push) Successful in 2s
CI / frontend-build (push) Successful in 21s
CI / backend-lint-and-test (push) Successful in 27s
CI / integration (push) Successful in 3m25s

- Migration 0069: new installs default to SigLIP 2 (so400m, 512px, 1152-d drop-in)
  — UPDATE applies ONLY where no image is embedded yet (fresh install), so an
  existing library is NOT silently invalidated; it switches deliberately via the
  dropdown → Re-embed → Retrain. Column server_defaults moved to SigLIP 2.
- GET /api/ml/embedder-models: server-authoritative supported list (SigLIP 2 512
  recommended / 384 faster / SigLIP 1 384 original) so the UI never free-types.
- GpuAgentCard: the two name/version text fields → a single model dropdown;
  Save sets name+version from the picked option (the current model is always
  selectable even if off-list).
- embedder.py DEFAULT_MODEL_NAME unchanged (stays the baked local-dir SigLIP 1)
  to avoid a local-dir/weights mismatch; SigLIP 2 loads by HF name, cached on the
  ml-worker's persistent HF_HOME.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
This commit is contained in:
2026-06-30 16:29:27 -04:00
parent 80f8eb4756
commit 359bc5a283
6 changed files with 133 additions and 31 deletions
+4 -2
View File
@@ -71,14 +71,16 @@ class MLSettings(Base):
ccip_auto_apply_threshold: Mapped[float] = mapped_column(
Float, nullable=False, default=0.92
)
# Default = SigLIP 2 (so400m, 512px) for new installs (migration 0069);
# existing libraries keep their stored value until the operator re-embeds.
embedder_model_version: Mapped[str] = mapped_column(
String(128), nullable=False, default="siglip-so400m-patch14-384"
String(128), nullable=False, default="siglip2-so400m-patch16-512"
)
# The HF model NAME the embedder loads (server CPU embed + announced to the
# GPU agent in the lease). Operator-settable so the embedder is a choice, not
# a hardcode (#1190): set name + version together, then re-embed + retrain.
embedder_model_name: Mapped[str] = mapped_column(
String(128), nullable=False, default="google/siglip-so400m-patch14-384"
String(128), nullable=False, default="google/siglip2-so400m-patch16-512"
)
updated_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), nullable=False, server_default=func.now()