opensearch-neural-sparse-encoding-multilingual-v1 · GGUF
GGUF conversion of opensearch-project/opensearch-neural-sparse-encoding-multilingual-v1
(168M-parameter inference-free multilingual sparse retrieval encoder, 105,879-token vocabulary), packaged for sqlite-sparse:
semantic search in a single SQLite file with no model at query time. Served as the
multilingual tier.
llama.cpp's converter drops the masked-LM head that sparse encoders need, so this repo
ships it as a sidecar: the .sprs file carries the MLM transform, the query-weight
lookup table (the "inference-free" query side), and the WordPiece vocabulary.
Files
| file | what |
|---|---|
multilingual_f16.gguf |
encoder body, f16 (llama.cpp-loadable, per-token hidden states) |
multilingual_q8.gguf |
encoder body, Q8_0 |
multilingual.sprs |
MLM head + query weight table + vocab (format v1) |
Fidelity
Differentially validated against the original SentenceTransformers implementation:
- encoder hidden states, worst token: cosine 0.999996 (f16), 0.999702 (Q8_0)
- end-to-end term weights through the sidecar head: max delta 1.3e-03
Validation probes included German and Spanish sentences with exact token-count agreement between llama.cpp and HF tokenization.
Provenance
Weights are unmodified from the Apache-2.0 original by the OpenSearch project; this repo only changes container format. All retrieval-quality credit belongs to the original model.