Aurekai

Try a model

Nomic static embedding, 128 dimensions

Static token embedding distilled by bonfyre-embed distill from the local nomic-embed-text-onnx graph (PCA to 128 dimensions, Zipf weighting, int8 rows). Static semantic search reads it in CMS publishing, the web address server and the browser. This identity is not the teacher's sentence model and implies no retrieval quality; what has been measured is carried by its representation passport.

Your input

Checking whether you can run this model…

Answer

The answer appears here. Nothing runs until you press Run.

About this model

Format
bfse
Size
4 MB
Shape
128-dimensional normalized static embedding
Family
none
Tested for embedding
measured, on hold

nomic-static-128/bfse-int8/2026-10-17. Teacher agreement measured, not retrieval quality. On all 1703 site search passages (corpus sha256 e3b2a8be) the .bfse shares 18.5 percent of each text's top-10 neighbours with the nomic-embed-text-onnx teacher (top-1 agreement 17.1 percent, worst query 0 of 10). On the first 2000 visible publication titles (sha256 9a8dc282, 55 refused as invisible) it shares 20.7 percent (top-1 22.0 percent; receipt sha256 47b89764). The distill receipt records 68.6 percent explained variance of token rows. Static embedding cost 0.0026 ms per text against 62 to 203 ms for the teacher on this Mac. No relevance judgement, false-prune trial or cross-model comparability has been measured; ResourceAdmission must not treat this representation as a substitute for its teacher or as a ranker of record. Check code 3810e4edffb745a7fc3a2f1842a2651daf4ead1ffd34de177db048edce4a641c.

Price
No price published

No provider lists a price for this model here. That doesn’t make it free to run.

Where to get it

An address recorded here, not a promise that it answers.

Use it in your work

Being listed here doesn’t mean you can run it. Running needs the model installed, access for your account and a limit you check before it starts.