virgil-meeting-e4b-a_v41

Meeting-synthesis split-LoRA bundle for the Virgil desktop in-process engine.

  • adapters/primary โ€” EXTRACT, A_v41: trained on the KV-prefix-cache-friendly prompt order (participants, recent_context, open_state_items, writable_new_chunk). A_v39's exact recipe; only the section layout differs. Pairs with render_extract_prompt as of runpod-jobs commit 68401685f โ€” renderer and adapter must ship together (either alone is off-format).
  • adapters/fallback โ€” RECONCILE, unchanged from the a_v32 bundle (a_v30 reconcile). The format change does not touch the reconcile prompt.
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for lightsofapollo/virgil-meeting-e4b-a_v41

Finetuned
(84)
this model