Nomic Embed
Nomic AINomic Embed produces long-context text embeddings for search and retrieval-augmented generation. The v1 and v1.5 encoders accept up to 8,192 tokens, and v1.5 is trained with Matryoshka representations so a vector can be truncated to 64 dimensions with graded loss of quality. Version 2 is a mixture-of-experts encoder covering roughly 100 languages at a 512-token limit. Nomic AI publishes weights, training data and the contrastors training pipeline together.
Scored over the nomic-embed-text line only. The first-party GGUF conversions are the same checkpoints in another container and are not counted twice, and nomic-embed-code, nomic-embed-vision and modernbert-embed-base are separate releases. Verified 2026-09-11 via the Hugging Face model cards, the Hub API and the contrastors README.
Openness
5 high confidence- weights
- open(Apache-2.0 ungated checkpoints for v1, v1.5 and v2-moe)
- data
- open(the card states the training data is released in its entirety, served through the contrastors repository behind a free Atlas account)
- code
- open(contrastors ships MLM pretraining, contrastive pretraining and finetuning entry points with their configs, plus the scripts that generate the data)
- license
- Apache-2.0(OSI)
The full-release case this rung was written for: weights, corpus and the pipeline that consumed it, all under Apache-2.0, with the v2 paper stating the point of the release is reproducibility of the training run. The corpus is fetched from a Nomic bucket after registering for an Atlas account, which is friction at the download gate rather than a corpus withheld.
- https://huggingface.co/api/models/nomic-ai/nomic-embed-text-v1.5 recorded 2026-09-11
license: apache-2.0; gated: false; private: false; downloadable checkpoint files present.
- https://huggingface.co/api/models/nomic-ai/nomic-embed-text-v2-moe recorded 2026-09-11
license: apache-2.0; gated: false; private: false; the mixture-of-experts checkpoint ships under the same terms.
- https://huggingface.co/nomic-ai/nomic-embed-text-v1.5/raw/main/README.md recorded 2026-09-11
States "Training data to train the models is released in its entirety" and points at the contrastors repository, and describes the two-stage unsupervised then supervised contrastive pipeline.
- https://raw.githubusercontent.com/nomic-ai/contrastors/main/README.md recorded 2026-09-11
Documents data access to the nomic-embed-text-v1 dataset, and gives the deepspeed and torchrun commands with configs for MLM pretraining, contrastive pretraining and finetuning, plus the scripts that generate new data.
- https://huggingface.co/nomic-ai/nomic-embed-text-v2-moe/raw/main/README.md recorded 2026-09-11
Says "Fully Open-Source: Model weights, code, and training data (see code repo) released" and that all code, models and evaluation data are open-sourced for full reproducibility of the training pipeline.
Adoption
5 high confidence16,124,079 downloads in the trailing 30 days for nomic-embed-text-v1.5 alone, plus 3,623,688 for v1 and 2,193,696 for v2-moe. The Hugging Face universe sweep puts the three shipped nomic-embed-text checkpoints at 22,689,628 combined, excluding the first-party GGUF conversions of the same weights.
- https://huggingface.co/api/models/nomic-ai/nomic-embed-text-v1.5 recorded 2026-09-11
downloads: 16124079.
- https://huggingface.co/api/models/nomic-ai/nomic-embed-text-v1 recorded 2026-09-11
downloads: 3623688.
- https://huggingface.co/api/models/nomic-ai/nomic-embed-text-v2-moe recorded 2026-09-11
downloads: 2193696.
Capability
3 high confidenceSolid and widely used, and measurably behind the frontier encoders on the same leaderboard table - 62.28 against 64.38 for a model of similar size. The v2 mixture of experts claims parity only within its parameter class, and gives up the long-context window the v1 line was known for.
- https://huggingface.co/Alibaba-NLP/gte-reranker-modernbert-base/raw/main/README.md recorded 2026-09-11
MTEB leaderboard table lists nomic-embed-text-v1.5 at 768 dimensions, 8192 maximum sequence length and an MTEB English average of 62.28, against 64.38 for gte-modernbert-base.
- https://huggingface.co/nomic-ai/nomic-embed-text-v2-moe/raw/main/README.md recorded 2026-09-11
Claims state-of-the-art multilingual performance compared with ~300M parameter models on BEIR and MIRACL, and states a maximum sequence length of 512 tokens.
Verified 2026-09-11