Computational Model Studio · Model release
BGE Reranker v2 M3
Scores how relevant a candidate passage is to a query, supporting a second stage of text retrieval.
Developer: Beijing Academy of Artificial Intelligence · Last verified:
What it is designed to do
Score a query and candidate passage for relevance so search results can be reordered.
Documented uses: Reranking candidate passages after initial text retrieval.
Access and deployment
- BGE Reranker v2 M3 model files — available. Public publisher repository; review applicable terms before use. Artifact documentation Local hardware requirements have not been verified. The full artifact-specific rights document has not been separately reviewed; consult the publisher repository conditions.
Technical profile and review notes
- Construction
- learned
- Representation
- neural
- Normalized architecture tags
- No normalized architecture tag assigned
- Lifecycle
- Unknown from reviewed sources
Inclusion rationale: Adds direct query-passage ranking, absent from the embedding-heavy seed and distinct from encoding text as vectors.
Limitations
- A relevance score is relative to the supplied query and candidate set; it does not retrieve documents by itself.
Sources and related Studio records
Sources
Connections
- In areaComputational Models
- Supports functionRerank candidates
- ArtifactBGE Reranker v2 M3 model files