Model reference · open weights
nomic-embed-text is an open-weight embedding model from nomic-ai, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
nomic-embed-text-v1.5: Resizable Production Embeddings with Matryoshka Representation Learning Blog | Technical Report | AWS SageMaker | Nomic Platform Exciting Update!: nomic-embed-text-v1.5 is now multimodal! nomic-embed-vision-v1.5 is aligned to the embedding space of nomic-embed-text-v1.5, meaning any text embedding is multimodal! Usage Important: the text prompt must include a task instruction prefix, instructing the model which task is being performed. For example, if you are implementing a RAG application, you embed your documents as searchdocument: <text here and embed your user queries as searchquery: <text here. Notice: From transformers v5.5.0 and sentence transformers v5.3.0, trustremotecode=True will no longer be necessary. This will only be possible with the text-only series as of now. Task instruction prefixes searchdocument Purpose: embed texts as documents from a dataset This prefix is used for embedding texts as documents, for example as documents for a RAG index. searchquery Purpose: embed texts as questions to answer This prefix is used for embedding texts as questions that documents from a dataset could resolve, for example as queries to be answered by a RAG application. clustering Purpose: embed texts to group them into clusters This prefix is used for embedding texts in order to group them into clusters, discover common topics, or remove semantic duplicates. classification Purpose: embed texts to classify them This prefix is used for embedding texts into vectors that will be used as features for a classification model Sentence Transformers Transformers The model natively supports scaling of the sequence length past 2048 tokens. To do so, Transformers.js Nomic API The easiest way to use Nomic Embed is through the Nomic Embedding API. Generating embeddings with the nomic Python client is as easy as For more information, see the API reference Infinity Usage with Infinity. Adjusting Dimensionality nomic-embed-text-v1.5 is an improvement upon Nomic Embed that utilizes Matryoshka Representation Learning which gives developers the flexibility to trade off the embedding size for a negligible reduction in performance. Training Click the Nomic Atlas
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | nomic-ai |
|---|---|
| Type | Embedding models |
| Parameters (lead) | 137M |
| Context | 2k tokens |
| Variants | 2 |
| Runs with | sentence-transformers |
| Released | 2024-02-10 |
| Popularity | 16.6M downloads / month |
| Likes | 895 |
| Licence | Open weights |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
Benchmarks
As published on the model card — the maker's own numbers, not measured by AxForge.
| Task | Dataset | Metric | Score |
|---|---|---|---|
| Classification | MTEB AmazonCounterfactualClassification (en) | accuracy | 75.209 |
| Classification | MTEB AmazonCounterfactualClassification (en) | ap | 38.576 |
| Classification | MTEB AmazonCounterfactualClassification (en) | f1 | 69.356 |
| Classification | MTEB AmazonPolarityClassification | accuracy | 91.814 |
| Classification | MTEB AmazonPolarityClassification | ap | 88.652 |
| Classification | MTEB AmazonPolarityClassification | f1 | 91.804 |
| Classification | MTEB AmazonReviewsClassification (en) | accuracy | 47.162 |
| Classification | MTEB AmazonReviewsClassification (en) | f1 | 46.593 |
| Retrieval | MTEB ArguAna | map_at_1 | 24.253 |
| Retrieval | MTEB ArguAna | map_at_10 | 38.962 |
| Retrieval | MTEB ArguAna | map_at_100 | 40.081 |
| Retrieval | MTEB ArguAna | map_at_1000 | 40.089 |
| Retrieval | MTEB ArguAna | map_at_3 | 33.499 |
| Retrieval | MTEB ArguAna | map_at_5 | 36.351 |
| Retrieval | MTEB ArguAna | mrr_at_1 | 24.609 |
| Retrieval | MTEB ArguAna | mrr_at_10 | 39.099 |
| Retrieval | MTEB ArguAna | mrr_at_100 | 40.211 |
| Retrieval | MTEB ArguAna | mrr_at_1000 | 40.219 |
| Retrieval | MTEB ArguAna | mrr_at_3 | 33.677 |
| Retrieval | MTEB ArguAna | mrr_at_5 | 36.469 |
| Retrieval | MTEB ArguAna | ndcg_at_1 | 24.253 |
| Retrieval | MTEB ArguAna | ndcg_at_10 | 48.011 |
| Retrieval | MTEB ArguAna | ndcg_at_100 | 52.756 |
| Retrieval | MTEB ArguAna | ndcg_at_1000 | 52.965 |
Using it via the API
Once AxForge deploys nomic-embed-text for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (nomic-embed-text below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/embeddings \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"nomic-embed-text","input":"text to embed"}'
Licence
Open weights under apache-2.0 — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗
Explore