Encoder V1
Meta🇺🇸 USFrontier· Nov 12, 2026
Embeddingopen weights
MetaEncoder v1 produces embeddings for images, video, documents and text that suit search, clustering and comparison. Its decide function also answers closed-set questions about a state with probabilities.
Summary of the model card
- Modalities
- text → embedding
1 variant
| Standard | HF |
Sources
HuggingFaceDeutsch
MetaEncoder v1 erzeugt Embeddings fuer Bilder, Videos, Dokumente und Text, die sich fuer Suche, Clustering und Vergleich eignen. Die Funktion decide beantwortet zusaetzlich geschlossene Fragen zu einem Zustand mit Wahrscheinlichkeiten.
Provenance
- Detected
- Nov 12, 2026
- First source
- huggingface
- Description
- Summary of the model card
- Editorially reviewed
- —