State-of-the-Art Embedding Models for Every Use Case

Open source, high-performance embedding models for multilingual, multimodal, code, and document search applications.

Nomic Embed Text v2

State-of-the-art multilingual text embeddings

Nomic Embed Text v1.5

Efficient and performant english text embeddings

Nomic Embed Vision v1.5

Multimodal extension for Nomic Embed Text v1.5

Nomic Embed Code

State-of-the-art code embedding and reranking

Learn More

Why Choose Nomic Embeddings?

High Performance

Our models consistently outperform competitors on industry benchmarks like MTEB and MIRACL.

Open Source

Full access to model weights, training data, and code. No vendor lock-in or API dependencies.

Cost Efficient

Compact models with efficient inference. Run locally or in the cloud at a fraction of the cost.

Efficient Models. Uncompromising Performance.

Model	Avg	Bittext Mining	Class.	Clust.	Pair Class.	Reranking	Retrieval	STS	$/1M Token
Nomic Embed V2	62.48	65.12	60.37	45.59	76.33	61.72	57.26	71.03	$0.01
Voyage 3 Lite	60.88	60.12	57.93	45.69	75.07	60.29	58.92	68.20	$0.02
OpenAI Embed 3 Small	58.64	50.32	55.16	46.78	76.64	59.94	52.24	69.42	$0.02
Arctic Embed M 2.0	58.44	53.73	54.38	43.02	74.86	61.67	54.83	66.60	$0.01

Nomic Embed v2 provides best in class performance per dollar on the MMTEB benchmark.

Model	Avg	ar	bn	de	en	es	fa	fi	fr	hi	id	ja	ko	ru	sw	te	th	yo	zh
Nomic Embed v2	66.0	76.7	73.6	56.6	54.7	56.3	59.2	77.1	55.8	60.5	54.2	67.0	65.9	65.2	66.3	82.6	78.3	78.3	59.5
Voyage-3-Large	59.5	69.6	68.3	46.2	48.4	43.8	51.1	70.8	39.8	54.8	47.2	62.3	63.9	57.8	67.9	76.7	74.5	75.6	52.1
OpenAI Text Embedding 3 Large	54.9	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-

Nomic Embed v2 achieves state-of-the-art performance on the multilingual MIRACL benchmark.

Model	Python	Java	Ruby	PHP	JavaScript	Go
Nomic Embed Code	81.6	80.5	81.9	72.3	77.1	93.8
Voyage Code 3	80.9	80.5	84.6	71.7	79.2	93.2
Nomic CodeRankEmbed-137M	78.4	76.9	79.3	68.8	71.4	92.7
OpenAI Embed 3 Large	70.8	72.9	75.3	59.6	68.1	87.6
CodeSage Large v2	74.2	72.3	76.7	65.2	72.5	84.6

The Nomic Embed Code suite achieves state-of-the-art performance on the CodeSearchNet benchmark.

Model	Avg.	ESG Restaurant Human	Econ Macro Multi.	AXA Multi.	MIT Bio	ESG Restaurant Synth.	ESG Restaurant Synth. Multi.	MIT Bio Multi.	AXA	Econ. Macro
ColNomic Embed Multimodal 7B	62.7	73.9	54.7	61.3	66.1	57.3	56.7	64.2	68.3	61.6
ColNomic Embed Multimodal 3B	61.2	65.8	55.4	61.0	63.5	56.6	57.2	62.5	68.8	60.2
T-Systems ColQwen2.5-3B	59.9	72.1	51.2	60.0	65.3	51.7	53.3	61.7	69.3	54.8
Nomic Embed Multimodal 7B	59.7	65.7	57.7	59.3	64.0	49.2	51.9	61.2	66.3	63.1
GME Qwen2 7B	59.0	65.8	56.2	55.4	64.0	54.3	56.7	55.1	60.7	62.9
Nomic Embed Multimodal 3B	58.8	59.8	57.5	58.8	62.5	49.4	49.4	58.6	69.6	63.5
Llama Index vdr-2b-multi-v1	58.4	63.1	52.8	61.0	60.6	50.3	51.2	56.9	68.8	61.2
Voyage Multimodal 3	55.0	56.1	55.0	59.5	56.4	47.2	46.2	51.5	64.1	58.8
Amazon Titan Multimodal	20.3	18.6	20.6	14.2	33.9	8.5	10.1	22.6	21.7	32.6

Nomic Embed Multimodal delivers superior document embedding performance for PDF search and retrieval.

Learn More about Nomic Embed Benchmarks

The Developer's Choice

35M+

Downloads on Hugging Face

Benchmarks can be gamed. Usage can't. Join the 35+ Million developers who trust Nomic embedding models for their production applications.

Embedding Solutions for Any Application

The Nomic Embedding Ecosystem is designed to provide the best embedding models for any application. We offer a range of models for different use cases, including multilingual, multimodal, code, and document search applications.

Model	Description	Key Features
Nomic Embed v2	A state of the art multilingual embedder with a mixture-of-experts architecture.	•SOTA performance on the MIRACL benchmark •Support for 100+ languages •305M active parameters for efficient inference •Open weights, data, and code
Nomic Embed Multimodal	A state of the art multimodal embedder.	•SOTA performance on the Vidore Benchmark •Support for interleaved text and rich visual inputs such as PDFs, screenshots, •Open weights, data, and code
Nomic Embed Text v1.5	The most popular open source text embedder on Hugging Face.	•Outperforms OpenAI Embeddings on the MTEB benchmark •Matryoshka and Binary embeddings for efficient storage •137M parameters for efficient inference •Open weights, data, and code
Nomic Embed Vision v1.5	A vision embedder aligned to the Nomic Embed Text v1.5 latent space.	•Multimodal extension to Nomic Embed Text v1.5 •Strong performance across text, image, and mixed modality search in a unified latent space •Open weights, data, and code
Nomic Embed Code	A state of the art code embedding model.	•SOTA performance on the CodeSearchNet benchmark •Supports Python, Javascript, Java, Go, PHP, and Ruby •Pair with Nomic CodeRankEmbed-137M efficient inference •Open weights, data, and code

Nomic Embed v2

A state of the art multilingual embedder with a mixture-of-experts architecture.

Key Features:

•SOTA performance on the MIRACL benchmark
•Support for 100+ languages
•305M active parameters for efficient inference
•Open weights, data, and code

Nomic Embed Multimodal

A state of the art multimodal embedder.

Key Features:

•SOTA performance on the Vidore Benchmark
•Support for interleaved text and rich visual inputs such as PDFs, screenshots,
•Open weights, data, and code

Nomic Embed Text v1.5

The most popular open source text embedder on Hugging Face.

Key Features:

•Outperforms OpenAI Embeddings on the MTEB benchmark
•Matryoshka and Binary embeddings for efficient storage
•137M parameters for efficient inference
•Open weights, data, and code

Nomic Embed Vision v1.5

A vision embedder aligned to the Nomic Embed Text v1.5 latent space.

Key Features:

•Multimodal extension to Nomic Embed Text v1.5
•Strong performance across text, image, and mixed modality search in a unified latent space
•Open weights, data, and code

Nomic Embed Code

A state of the art code embedding model.

Key Features:

•SOTA performance on the CodeSearchNet benchmark
•Supports Python, Javascript, Java, Go, PHP, and Ruby
•Pair with Nomic CodeRankEmbed-137M efficient inference
•Open weights, data, and code

Atlas

GPT4All

Download Privacy Policy Terms of Service Learn More

Nomic

Careers

“Henceforth, it is the map that precedes the territory” – Jean Baudrillard