ModernBERT-base lineage
According to the model card, it is trained from ModernBERT-base, and hub tags record answerdotai/ModernBERT-base as the base model.
Open Source Model Profile · nomic-ai
modernbert-embed-base
modernbert-embed-base is a 149M-parameter ModernBERT embedding model from nomic-ai. The model card describes it as trained from ModernBERT-base with Matryoshka support.
- Publisher
- nomic-ai
- Task
- sentence-similarity
- Model type
- modernbert
- License
- apache-2.0
- Library
- sentence-transformers
- Publication status
- Accepted · not indexed
Model overview
modernbert-embed-base is published by nomic-ai as a sentence-similarity embedding model. The captured configuration identifies ModernBertModel with a modernbert model type and Safetensors metadata reports 149014272 parameters. According to the model card, it brings ModernBERT advances to embeddings and supports Matryoshka 256 dimensions.
Recorded capabilities
Matryoshka dimensions
The model card documents Matryoshka support down to 256 dimensions, with Sentence Transformers truncate_dim and transformers slicing examples.
Prefix-dependent inputs
According to the model card, inputs require Nomic-style prefixes such as search_query and search_document, with transformers 4.48.0 or newer.
Publisher-reported MTEB table
The model card includes an MTEB comparison reporting 62.62 average at 768 dimensions and 61.17 at 256 dimensions.
Use cases in the source record
- Retrieval-style embedding workflows using search_query and search_document prefixes in Sentence Transformers.
- Memory-efficient vector search using truncated 256-dimension Matryoshka embeddings.
Limitations and unknowns
- No independent Ethen evaluation results were extracted; MTEB figures come from the publisher model card.
- No context-window value was extracted from this record.
- Provider state is historical snapshot data, not independently refreshed current availability.
- Training-pipeline details come from the publisher model card and have not been independently verified by Ethen.
Source and provenance
Source: nomic-ai/modernbert-embed-base
Captured: Unknown. Processed: 2026-09-07T19:35:28.074739+00:00.
ModernBERT Embed ModernBERT Embed is an embedding model trained from ModernBERT-base , bringing the new advances of ModernBERT to embeddings! Trained on the Nomic Embed weakly-supervised and supervised datasets, modernbert-embed also supports Matryoshka Representation Learning dimensions of 256, reducing memory by 3x with minimal performance loss. Performance Model Dimensions Average (56) Classification (12) Clustering (11) Pair Classification (3) Reranking (4) Retrieval (15) STS (10) Summarization (1) nomic-embed-text-v1 768 62.4 74.1 43.9 85.2 55.7 52.8 82.1 30.1 nomic-embed-text-v1.5 768 62.28 73.55 43.93 84.61 55.78 53.01 81.94…
F001F002F003F004F005F006F007F009F010F011F012F013F014F015F016F017