Ollama (AnyRouter Relay)

BAAI's BGE-Large is an open multilingual text embedding model returning a 1,024-dimension dense vector over a 512-token window, across 100+ languages. MIT-licensed. AnyRouter exposes the text path via /embeddings on the user's own paired device running Ollama.

EmbeddingGemma 2 is Google DeepMind's open-weight multimodal embedding model with 740M parameters and one 768-dimensional vector space. This listing takes text input only, with an 8,192-token window.

Granite Embedding is IBM's open text embedding family in 30M and 278M sizes, licensed under Apache-2.0. It is built for retrieval, clustering and semantic search. The 278M model returns 768 dimensions and the 30M model returns fewer.

MX-BAI Embed Large is Mixedbread AI's open text embedding model, released under the MIT license. It returns a 1,024-dimension dense vector over a 512-token window. Input is text and output is embeddings.

Nomic Embed Text is an open text embedding model from Nomic AI, trained for semantic search and retrieval. The v1.5 cut returns a 768-dimension dense vector over an 8,192-token window and is MIT-licensed.

All MiniLM is the sentence-transformers MiniLM family (L6-v2 and L12-v1 cuts) for fast, compact text embedding. It returns a 384-dimension dense vector over a 256-token window and is MIT-licensed.

Snowflake Arctic Embed is Snowflake's open text embedding family, spanning small to large cuts across 100+ languages. Dimensions and context vary by tag, and arctic-embed-base returns 1,024 dimensions. It is Apache-2.0 licensed.

Strands Decider 2B is AWS Strands Labs' small open-weights (Apache 2.0) decision model. It takes application state and typed questions and returns calibrated probabilities in one forward pass, not generated chat text. Weights run on your own hardware.